What do you think?
Rate this book
On Intelligence: How a New Understanding of the Brain Will Lead to the Creation of Truly Intelligent Machines
From the inventor of the PalmPilot comes a new and compelling theory of intelligence, brain function, and the future of intelligent machines
Jeff Hawkins, the man who created the PalmPilot, Treo smart phone, and other handheld devices, has reshaped our relationship to computers. Now he stands ready to revolutionize both neuroscience and computing in one stroke, with a new understanding of intelligence itself.
Hawkins develops a powerful theory of how the human brain works, explaining why computers are not intelligent and how, based on this new theory, we can finally build intelligent machines.
The brain is not a computer, but a memory system that stores experiences in a way that reflects the true structure of the world, remembering sequences of events and their nested relationships and making predictions based on those memories. It is this memory-prediction system that forms the basis of intelligence, perception, creativity, and even consciousness.
In an engaging style that will captivate audiences from the merely curious to the professional scientist, Hawkins shows how a clear understanding of how the brain works will make it possible for us to build intelligent machines, in silicon, that will exceed our human ability in surprising ways.
Written with acclaimed science writer Sandra Blakeslee, On Intelligence promises to completely transfigure the possibilities of the technology age. It is a landmark book in its scope and clarity.
Jeff Hawkins, the man who created the PalmPilot, Treo smart phone, and other handheld devices, has reshaped our relationship to computers. Now he stands ready to revolutionize both neuroscience and computing in one stroke, with a new understanding of intelligence itself.
Hawkins develops a powerful theory of how the human brain works, explaining why computers are not intelligent and how, based on this new theory, we can finally build intelligent machines.
The brain is not a computer, but a memory system that stores experiences in a way that reflects the true structure of the world, remembering sequences of events and their nested relationships and making predictions based on those memories. It is this memory-prediction system that forms the basis of intelligence, perception, creativity, and even consciousness.
In an engaging style that will captivate audiences from the merely curious to the professional scientist, Hawkins shows how a clear understanding of how the brain works will make it possible for us to build intelligent machines, in silicon, that will exceed our human ability in surprising ways.
Written with acclaimed science writer Sandra Blakeslee, On Intelligence promises to completely transfigure the possibilities of the technology age. It is a landmark book in its scope and clarity.
272 pages, Hardcover
First published January 1, 2004
652 people are currently reading
12872 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 30 of 506 reviews
July 9, 2013
Okay. This book and I didn't get along terribly well, but the experience was nevertheless a valuable one. So, 3 stars, even though I disagree fundamentally with some of the theory and the style of presentation. This will be a long one; bear with me.
To put it simply.... Jeff Hawkins is a very intelligent computer engineer who thinks he understands brains in ways that no neuroscientist ever has before, mostly because he is willing to stand by a grand picture where most neuroscientists want to investigate every small chunk before declaring they've solved brains. He has read a lot of books about neuroscience and has spoken to a lot of neuroscientists, and has trudged up a (not patently incorrect) theory from the 1970's and used it as a foundation for what he considers the first general theory of cortical brain functioning (it isn't). He then equates "cortex" with "intelligence" and takes off on a grand tour of his theory: that we can build intelligent* machines to perform complex pattern recognition tasks in much the same ways that he proposes an organic cortex does.
*NOT human-like, though you wouldn't know it from the wording in any blurb you read about the book, including the book's own jacket summary.
There are a lot of theoretic assumptions in this book, and unpacking them is quite unfortunately left up to the reader, who may not have the requisite background knowledge to separate out responsible assumptions, backed by data that Hawkins rarely mentions in order to keep it digestible to a lay audience, from irresponsible ones. There are many of the latter, detectable only by those who know the field or the scientific method well enough to know a red flag when they see it. Hawkins plays loose on owning up to these assumptions, even when they are cornerstones without which his theory loses a lot of its appeal. I was relieved when he admitted to the oversimplification of his view at the end of the chapter on neurology, but it felt like an afterthought that could (should?) have been used to temper his conviction and factual flippancy up to that point. The tone of the book is occasionally that of a conspiracy theorist who has figured it all out, against all conventional belief, and pulls you along a fast sequence of premises and conclusions while waving his hands and telling you the details of the premises are too complicated to get into.
One of the bafflers that stood out to me (this is a bit technical, sorry) was the notion that the basal ganglia and cerebellum are old structures whose functions have been largely subsumed by the neocortex, and thus were unnecessary for a theory of intelligent motor behaviour. The balls it takes to make that claim, when such vastly debilitating diseases as Parkinson's, Huntingtons, or Ataxia exist, blew my brains out a bit.
At that point, it became clear that Hawkins was so fixated on the neocortex that he was willing to push aside contradictory evidence from subcortical structures to make his theory fit. I've seen this before, from neuroscientists who fall in love with a given brain region and begin seeing it as the root of all behaviour, increasingly neglecting the quite patent reality of an immensely distributed system. It's pretty natural, and honestly not limited to neuroscientists: when you stare too closely at one piece of a puzzle, you begin to forget that there are other pieces. For most scientists, however, this view need not be a detriment, because they generally aim their research programs at very specific questions-- questions that this atomized focus are fit to answer. In the case of Jeff Hawkins' general theory of brain function, however, it's entirely disingenuous.
Putting aside my qualms with his approach to the theory, there is actually some overlap between his views my own, and some points of valuable and probably instructive disagreement. Hawkins views intelligence as a result of hierarchical and recursive neural organization: basically, there are higher and lower levels, with communication tracing both upwards (from sensory input) and downwards (from higher levels of analysis) via patterns of activation. What we experience is a complex interaction between external input and internal input from memory, resulting in a continuous stream of online prediction. Up to this point, my theories of what we might call consciousness match his theory of intelligent pretty well. Where we differ is in the details (and in our respective convictions that we are correct!).
In neuroscience, there is a theoretical construct called the 'grandmother cell,' to illustrate the ludicrous idea that there is a single neuron in your brain that represents your grandmother, another for your cat, and so on. The grandmother cell has been disproven time and time again: the brain is a HIGHLY distributed system, and a given representation is the result of patterns of activations across many cells, not one cell. Jeff Hawkins acknowledges this.... before proposing instead that representations in the cortex are handled by (my term, not his) grandmother cellular columns. Briefly, the visual cortex has been shown to have a columnar organization that traverses six parallel layers of anatomically and biologically different cell types, such that cells at Point X of layer 1 respond to the same sorts of basic visual information (e.g., line angles) as cells at Point X of layers 2 through 6. Because the rest of the cortex also seems to be organized in six distinguishable layers, Hawkins suggests that the entire cortex operates in columns, such that the composite idea of your grandmother should be represented by a given cellular column in a high-level area of cortex. He never states these logical conclusions outright, but they follow from the way his theory proposes hierarchical organization to work. He briefly admits the oversimplification at play, and then nevertheless uses the oversimplification as the foundation for the rest of the theory. This is not a novel theory so much as it is an outdated theory with the goalpost pushed back one step. And while oversimplifications are a necessary evil in scientific progress*, they need to be acknowledged and admitted so that they can be refined, again, especially where a general theory for a lay audience is the goal.
*as my advisor says it, the goal of a scientist is to maintain a productive level of ambiguity.
The rest of the book was (to me) less controversial. There was the requisite chapter to answer such questions as "does this mean animals are intelligent!?" for readers who've never thought about the implications of a physical and evolutionarily-derived brain before. Yawn. This was followed by a chapter on what I assume is the whole reason Hawkins wrote the book: the prospect of intelligent machines.
Having defined 'intelligence' as 'cortex,' he rather plainly announces that an intelligent machine will be one organized with recursive and flexible hierarchies, a reveal that will shock or excite no cognitive scientist. He very clearly explains why current artificial intelligence built on existing computer memory structures are not up to the task, an argument that AI researchers have been ignoring for decades. Much to my pleasant surprise, again given the blurbs on this book, he then laments the cold hard reality that we will never have viable machines that are intelligent in the way humans are: humans are intelligent the way humans are because of all the sensory and proprioceptive input coming in from their human bodies to shape their brains. Unless we build almost impossibly costly and cumbersome human-like bodies to go with our fancy intelligent machine brains, it's a moot point trying to make machines like humans-- and why would we want to anyways? Hawkins outlines some realistic goals that are achievable (e.g., self-driving cars, diagnostic machines, weather prediction machines...), none of which I particularly disagree with except for the optimistic time-frames forecasted.
However, I can't help feeling that most readers are set up to be vastly disappointed by the propositions. With the majority of the book devoted to neurological theory, it's hard not to anticipate that the machine intelligence he will eventually propose will mimic neurology. The book jacket itself claims that Hawkins' theories will "make it possible for us to build intelligent machines, in silicon, that will not simply imitate but exceed our human ability in surprising ways." But nothing about the analogical applications of his neurologically-based theory are intended to imitate humans. He expressly states, in fact, that the aim to build artificial humans is wrong-footed and fated to fail. The message is a bit confusing, and while I would have personally been offended to hear him say what most readers likely wanted to hear-- that we can build human-like machines by analogy to human cortex-- I again get a sticky sense of disingenuity, this time to sell book copies.
Overall, this was an interesting but infuriating book that takes some great ideas from existing cognitive science, laudably exposes them to a lay audience in ways that most cognitive and neuroscientists won't bother to, shoves them into a flawed neurological framework, and then announces brains to be solved. The ego involved is staggering, the conclusions less so, and the applications underwhelming. I am admittedly very interested to see, hopefully in my lifetime, just how intelligent Hawkins' intelligent machines can get with only an analogical neocortex. Since he never discusses this fact, spoilers: a neocortex is not enough for either humans or nonhuman animals to function, let alone intelligently. The (rather expensive...) exercise of trying to evoke intelligence from cortex alone could provide us with a better appreciation of subcortical structures, much needed in this species-self-congratulatory era of cortical fixation.
To put it simply.... Jeff Hawkins is a very intelligent computer engineer who thinks he understands brains in ways that no neuroscientist ever has before, mostly because he is willing to stand by a grand picture where most neuroscientists want to investigate every small chunk before declaring they've solved brains. He has read a lot of books about neuroscience and has spoken to a lot of neuroscientists, and has trudged up a (not patently incorrect) theory from the 1970's and used it as a foundation for what he considers the first general theory of cortical brain functioning (it isn't). He then equates "cortex" with "intelligence" and takes off on a grand tour of his theory: that we can build intelligent* machines to perform complex pattern recognition tasks in much the same ways that he proposes an organic cortex does.
*NOT human-like, though you wouldn't know it from the wording in any blurb you read about the book, including the book's own jacket summary.
There are a lot of theoretic assumptions in this book, and unpacking them is quite unfortunately left up to the reader, who may not have the requisite background knowledge to separate out responsible assumptions, backed by data that Hawkins rarely mentions in order to keep it digestible to a lay audience, from irresponsible ones. There are many of the latter, detectable only by those who know the field or the scientific method well enough to know a red flag when they see it. Hawkins plays loose on owning up to these assumptions, even when they are cornerstones without which his theory loses a lot of its appeal. I was relieved when he admitted to the oversimplification of his view at the end of the chapter on neurology, but it felt like an afterthought that could (should?) have been used to temper his conviction and factual flippancy up to that point. The tone of the book is occasionally that of a conspiracy theorist who has figured it all out, against all conventional belief, and pulls you along a fast sequence of premises and conclusions while waving his hands and telling you the details of the premises are too complicated to get into.
One of the bafflers that stood out to me (this is a bit technical, sorry) was the notion that the basal ganglia and cerebellum are old structures whose functions have been largely subsumed by the neocortex, and thus were unnecessary for a theory of intelligent motor behaviour. The balls it takes to make that claim, when such vastly debilitating diseases as Parkinson's, Huntingtons, or Ataxia exist, blew my brains out a bit.
At that point, it became clear that Hawkins was so fixated on the neocortex that he was willing to push aside contradictory evidence from subcortical structures to make his theory fit. I've seen this before, from neuroscientists who fall in love with a given brain region and begin seeing it as the root of all behaviour, increasingly neglecting the quite patent reality of an immensely distributed system. It's pretty natural, and honestly not limited to neuroscientists: when you stare too closely at one piece of a puzzle, you begin to forget that there are other pieces. For most scientists, however, this view need not be a detriment, because they generally aim their research programs at very specific questions-- questions that this atomized focus are fit to answer. In the case of Jeff Hawkins' general theory of brain function, however, it's entirely disingenuous.
Putting aside my qualms with his approach to the theory, there is actually some overlap between his views my own, and some points of valuable and probably instructive disagreement. Hawkins views intelligence as a result of hierarchical and recursive neural organization: basically, there are higher and lower levels, with communication tracing both upwards (from sensory input) and downwards (from higher levels of analysis) via patterns of activation. What we experience is a complex interaction between external input and internal input from memory, resulting in a continuous stream of online prediction. Up to this point, my theories of what we might call consciousness match his theory of intelligent pretty well. Where we differ is in the details (and in our respective convictions that we are correct!).
In neuroscience, there is a theoretical construct called the 'grandmother cell,' to illustrate the ludicrous idea that there is a single neuron in your brain that represents your grandmother, another for your cat, and so on. The grandmother cell has been disproven time and time again: the brain is a HIGHLY distributed system, and a given representation is the result of patterns of activations across many cells, not one cell. Jeff Hawkins acknowledges this.... before proposing instead that representations in the cortex are handled by (my term, not his) grandmother cellular columns. Briefly, the visual cortex has been shown to have a columnar organization that traverses six parallel layers of anatomically and biologically different cell types, such that cells at Point X of layer 1 respond to the same sorts of basic visual information (e.g., line angles) as cells at Point X of layers 2 through 6. Because the rest of the cortex also seems to be organized in six distinguishable layers, Hawkins suggests that the entire cortex operates in columns, such that the composite idea of your grandmother should be represented by a given cellular column in a high-level area of cortex. He never states these logical conclusions outright, but they follow from the way his theory proposes hierarchical organization to work. He briefly admits the oversimplification at play, and then nevertheless uses the oversimplification as the foundation for the rest of the theory. This is not a novel theory so much as it is an outdated theory with the goalpost pushed back one step. And while oversimplifications are a necessary evil in scientific progress*, they need to be acknowledged and admitted so that they can be refined, again, especially where a general theory for a lay audience is the goal.
*as my advisor says it, the goal of a scientist is to maintain a productive level of ambiguity.
The rest of the book was (to me) less controversial. There was the requisite chapter to answer such questions as "does this mean animals are intelligent!?" for readers who've never thought about the implications of a physical and evolutionarily-derived brain before. Yawn. This was followed by a chapter on what I assume is the whole reason Hawkins wrote the book: the prospect of intelligent machines.
Having defined 'intelligence' as 'cortex,' he rather plainly announces that an intelligent machine will be one organized with recursive and flexible hierarchies, a reveal that will shock or excite no cognitive scientist. He very clearly explains why current artificial intelligence built on existing computer memory structures are not up to the task, an argument that AI researchers have been ignoring for decades. Much to my pleasant surprise, again given the blurbs on this book, he then laments the cold hard reality that we will never have viable machines that are intelligent in the way humans are: humans are intelligent the way humans are because of all the sensory and proprioceptive input coming in from their human bodies to shape their brains. Unless we build almost impossibly costly and cumbersome human-like bodies to go with our fancy intelligent machine brains, it's a moot point trying to make machines like humans-- and why would we want to anyways? Hawkins outlines some realistic goals that are achievable (e.g., self-driving cars, diagnostic machines, weather prediction machines...), none of which I particularly disagree with except for the optimistic time-frames forecasted.
However, I can't help feeling that most readers are set up to be vastly disappointed by the propositions. With the majority of the book devoted to neurological theory, it's hard not to anticipate that the machine intelligence he will eventually propose will mimic neurology. The book jacket itself claims that Hawkins' theories will "make it possible for us to build intelligent machines, in silicon, that will not simply imitate but exceed our human ability in surprising ways." But nothing about the analogical applications of his neurologically-based theory are intended to imitate humans. He expressly states, in fact, that the aim to build artificial humans is wrong-footed and fated to fail. The message is a bit confusing, and while I would have personally been offended to hear him say what most readers likely wanted to hear-- that we can build human-like machines by analogy to human cortex-- I again get a sticky sense of disingenuity, this time to sell book copies.
Overall, this was an interesting but infuriating book that takes some great ideas from existing cognitive science, laudably exposes them to a lay audience in ways that most cognitive and neuroscientists won't bother to, shoves them into a flawed neurological framework, and then announces brains to be solved. The ego involved is staggering, the conclusions less so, and the applications underwhelming. I am admittedly very interested to see, hopefully in my lifetime, just how intelligent Hawkins' intelligent machines can get with only an analogical neocortex. Since he never discusses this fact, spoilers: a neocortex is not enough for either humans or nonhuman animals to function, let alone intelligently. The (rather expensive...) exercise of trying to evoke intelligence from cortex alone could provide us with a better appreciation of subcortical structures, much needed in this species-self-congratulatory era of cortical fixation.
April 1, 2009
"On Intelligence (and Condescension)"
The only thing wrong with Jeff Hawkins's book is Jeff Hawkins. His idea for the brain basic structure is exciting. (basically he argues that the brain works off a near-recursive prediction model based on stimulus and memory.) And he's really into intelligent machines. In fact, he may have convinced me not to fear the giant robot armies that have plagued my dreams. Nope. Now I can fear the infinitely-sized hyper-conscious EverMind that operates mainly in the sixth dimension.
But don't let that sort of thing keep you away. He doesn't get to that until chapter 8. (ish).
Unfortunately, like a great many successful industrialists who have spent most of their life outside of academia (I'm talking to you, Mr. Soros), Hawkins writes with a chip on his shoulder and something to prove. Mainly that he's smart. Really really really smart. It would have been a far nicer read if he'd stuck to proving his thesis.
Notice in the beginning how he goes out of his way to point out that Sandra Blakeslee (science writer) is NOT the author...just a ghost writer or editor or something like that. It's belittling and a little embarrassing. You'd think a lifetime student of the mind would be less inclined to pride.
Okay. If you want to learn more about the brain than you know, probably, then this is a great book to read. But if blowhards bother you, maybe you should reconsider your curiosity. It is an unfortunate fact that a lot of these industrialists have really great ideas and so are forced to promote them in these kind of books. (Now I'm talking to you Mr. Wolfram). The capitalists have no humility, but are often smart and right.
5 stars for content
1 star for voice
3 stars and praise for the book this could have been.
The only thing wrong with Jeff Hawkins's book is Jeff Hawkins. His idea for the brain basic structure is exciting. (basically he argues that the brain works off a near-recursive prediction model based on stimulus and memory.) And he's really into intelligent machines. In fact, he may have convinced me not to fear the giant robot armies that have plagued my dreams. Nope. Now I can fear the infinitely-sized hyper-conscious EverMind that operates mainly in the sixth dimension.
But don't let that sort of thing keep you away. He doesn't get to that until chapter 8. (ish).
Unfortunately, like a great many successful industrialists who have spent most of their life outside of academia (I'm talking to you, Mr. Soros), Hawkins writes with a chip on his shoulder and something to prove. Mainly that he's smart. Really really really smart. It would have been a far nicer read if he'd stuck to proving his thesis.
Notice in the beginning how he goes out of his way to point out that Sandra Blakeslee (science writer) is NOT the author...just a ghost writer or editor or something like that. It's belittling and a little embarrassing. You'd think a lifetime student of the mind would be less inclined to pride.
Okay. If you want to learn more about the brain than you know, probably, then this is a great book to read. But if blowhards bother you, maybe you should reconsider your curiosity. It is an unfortunate fact that a lot of these industrialists have really great ideas and so are forced to promote them in these kind of books. (Now I'm talking to you Mr. Wolfram). The capitalists have no humility, but are often smart and right.
5 stars for content
1 star for voice
3 stars and praise for the book this could have been.
October 8, 2012
I liked this book: it contains a few nice thought experiments about intelligence. Bare in mind, however, that Jeff Hawkins' implementation of these ideas has not proven to be fruitful so far in his company Numenta.
July 14, 2008
Hawkins' theory is that the entire sensory cortex runs a single cortical algorithm to perform all of its sensory functions.
This single algorithm simply looks for patterns. Layers and layers of brain cells performing this pattern recognition result in our sensory experience. Here is an example of how this might work for vision:
Layer 1 receives sensory input from the outside world and looks for general patterns of lines.
Layer 2 receives input from layer 1 and looks for patterns of edges from those lines.
Layer 3 receives input from layer 2 and looks for groups of edges forming rectangles and other shapes.
Layer 4 receives input from layer 3 and looks for patterns of shapes (a circle on top of a vertical rectangle with 2 long vertical rectangles below and 2 long horizontal rectangles on the side might be a human with 4 limbs).
...and on and on
*I'm not a neuroscientist. Go read the book to get a much better explanation of this theory.
The variety of sensory functions our brain can perform then arise, not from a set of varied algorithms, but from the structure of the brain.
Hawkins also suggests that behavior and prediction are one and the same. To perform an action, our brains imagine us going through the action, our motor neurons fire in response, and voila!, we act. (Its been a few months since I read this book and my memory on this theory is a bit fuzzy. Again, go read the book.)
The great leaps forward in science of our past have often been counterintuitive simplifications of a complex issue and Hawkins' theory is just simple and elegant enough to qualify. Research over the next decade will prove him right or wrong, but in either case, this is a fantastic book to stimulate thinking on what's going on inside your head.
(Note: this is a scientifically dense book and you may have to reread sentences, paragraphs, and chapters to make sense of it all. I personally intend to go back and read it again after reading a few other books on intelligence and the brain.)
This single algorithm simply looks for patterns. Layers and layers of brain cells performing this pattern recognition result in our sensory experience. Here is an example of how this might work for vision:
Layer 1 receives sensory input from the outside world and looks for general patterns of lines.
Layer 2 receives input from layer 1 and looks for patterns of edges from those lines.
Layer 3 receives input from layer 2 and looks for groups of edges forming rectangles and other shapes.
Layer 4 receives input from layer 3 and looks for patterns of shapes (a circle on top of a vertical rectangle with 2 long vertical rectangles below and 2 long horizontal rectangles on the side might be a human with 4 limbs).
...and on and on
*I'm not a neuroscientist. Go read the book to get a much better explanation of this theory.
The variety of sensory functions our brain can perform then arise, not from a set of varied algorithms, but from the structure of the brain.
Hawkins also suggests that behavior and prediction are one and the same. To perform an action, our brains imagine us going through the action, our motor neurons fire in response, and voila!, we act. (Its been a few months since I read this book and my memory on this theory is a bit fuzzy. Again, go read the book.)
The great leaps forward in science of our past have often been counterintuitive simplifications of a complex issue and Hawkins' theory is just simple and elegant enough to qualify. Research over the next decade will prove him right or wrong, but in either case, this is a fantastic book to stimulate thinking on what's going on inside your head.
(Note: this is a scientifically dense book and you may have to reread sentences, paragraphs, and chapters to make sense of it all. I personally intend to go back and read it again after reading a few other books on intelligence and the brain.)
July 27, 2011
Jeff Hawkins has done a remarkable thing. He's essentially synthesised all of the information we have on how the brain works into a simple, elegant and utterly comprehensible theory of intelligence that will pave the way to the creation of truly intelligent machines. That's a massive claim I know but I honestly don't think I have ever read a simpler, more straightforward account of what intelligence is.
Hawkins' theory, in a nutshell, is that intelligence is a manifestation of the brains ability to predict the future and test its perceptions against its predictions. Like a fractal there is a mass of self-similarity at work here. At the very fine-grained level the predictions the brains making are very mundane but as sensory information is handled, and exceptions passed up the hierarchy, and predictions passed back down the hierarchy, our brains learn from their experiences and, over time a genuine, common understanding of the world emerges.
Anyone working on machine intelligence should read this short, simple book.
Hawkins' theory, in a nutshell, is that intelligence is a manifestation of the brains ability to predict the future and test its perceptions against its predictions. Like a fractal there is a mass of self-similarity at work here. At the very fine-grained level the predictions the brains making are very mundane but as sensory information is handled, and exceptions passed up the hierarchy, and predictions passed back down the hierarchy, our brains learn from their experiences and, over time a genuine, common understanding of the world emerges.
Anyone working on machine intelligence should read this short, simple book.
August 7, 2013
I picked this up on sale but I can't finish it. I wanted a bright person's coherent and logical progression through a model of the brain. As bright as the author might be, he is astonishingly tone deaf to how distracting the relentless implicit and direct accolades he gives to himself are to the science he is trying to explain. The book might be summarized as a tapestry of ....introduction (all about me!)...look at me again!...science...look at me!...look at me!....science...did you see me?! etc
Assumptions and assertions about research threads other than his own are conveniently packaged with plenty of straw men, when with a little bit more scientific humility he could be so much more effective. I would love to see what he put in his later chapters, but can't deal with all the sludge you have to put up with to get the good stuff.
Assumptions and assertions about research threads other than his own are conveniently packaged with plenty of straw men, when with a little bit more scientific humility he could be so much more effective. I would love to see what he put in his later chapters, but can't deal with all the sludge you have to put up with to get the good stuff.
September 1, 2015
Moderately entertaining speculation on how intelligence works on a neural level in humans. A lot of his criticisms of neuroscience as practiced rung true to me (a more-than-layman less-than-initiate for this field) 10 years later, though I wouldn't be as extreme as he is. The framework he puts forth is at least plausible and has a certain elegance to it. Unfortunately, there's not a whole lot of support provided, and a lot of the assertions he makes (particularly his high confidence in Mountcastle's "Organizing Principle", which underlies the entire framework) are much stronger than seems justifiable, even given his explicit hedging. Given the core ideas around which his theory was built, I think Hawkins would have been much better served doing some cheap tests (literature review + novel experiments) to see if they can be easily discarded and then gone on to design experiments to flesh the theory out and validate it at the same time, rather than trying to build a full-blown framework and publishing a book.
The discussions around intelligent machines were interesting and raised some good points about how the AI field has progressed and what we might expect artificial intelligences to look like. I especially liked his dismissal of the behavior-intelligence equation and his avoiding the trap of "Church-Turing thesis, therefore von Neumann architecture is as good a choice as any" that so many seem to fall into. Ultimately, though, a lot of his speculation seemed really weak and more importantly it seemed remarkably premature to start speculative engineering before we have the science. I wouldn't put much stock in any specific predictions he makes either way. He also seriously strawmans the AI risk concern.
Finally, the book was hilarious in the "laugh at him, not with him" way. In addition to the story of the memory-prediction model, On Intelligence is also the story of how awesome Jeff Hawkins is and how silly everyone else is. The book is littered with spurious anecdotes, personal perspectives, and seemingly sour grapes at how misunderstood his genius is. I enjoyed this aspect of the book, but not in a way I'd expect Hawkins himself to appreciate.
The discussions around intelligent machines were interesting and raised some good points about how the AI field has progressed and what we might expect artificial intelligences to look like. I especially liked his dismissal of the behavior-intelligence equation and his avoiding the trap of "Church-Turing thesis, therefore von Neumann architecture is as good a choice as any" that so many seem to fall into. Ultimately, though, a lot of his speculation seemed really weak and more importantly it seemed remarkably premature to start speculative engineering before we have the science. I wouldn't put much stock in any specific predictions he makes either way. He also seriously strawmans the AI risk concern.
Finally, the book was hilarious in the "laugh at him, not with him" way. In addition to the story of the memory-prediction model, On Intelligence is also the story of how awesome Jeff Hawkins is and how silly everyone else is. The book is littered with spurious anecdotes, personal perspectives, and seemingly sour grapes at how misunderstood his genius is. I enjoyed this aspect of the book, but not in a way I'd expect Hawkins himself to appreciate.
September 3, 2019
জেফ্ হকিনসের সাধ ছিল বুদ্ধিমান যন্ত্র বানাবেন। বিহেভেরিয়রিজমের নিরিখে পাওয়া হাল আমলের বুদ্ধিমান যন্ত্র না, সত্যিকারের বুদ্ধিমান যন্ত্র যা তথ্য শুধু ব্যবহার না, বুঝতে সক্ষম। কিন্তু বুদ্ধিমান যন্ত্র বানাতে হলে বুদ্ধি কী তা তো আগে বোঝা দরকার। সেই জন্য তিনি নিউরোসায়েন্সের দ্বারস্থ হয়েছিলেন।
ফলাফল: অন ইনটেলিজেন্স।
মানুষের বুদ্ধির (এবং ভাষার) বিকাশ সম্পর্কিত জনপ্রিয় ন্যারেটিভটি কগনিটিভ রেভলিউশনের (Yuval Noah Harari-র কগনিটিভি রেভলিউশন, সোশ্যাল মুভমেন্ট না।) ওপর দাঁড়ানো। পৃথিবীর প্রাণীজগতের সামগ্রিক বিবর্তনের ল্যান্ডস্কেপে এই ন্যারেটিভটি প্রায় মাটি থেকে মানুষ তৈরীর মতই খাপ ছাড়া। সম্প্রতি বুদ্ধিমত্তা ও ভাষার বিকাশের বিবর্তনীয় ব্যাখ্যাগুলো পরিচিতি পাচ্ছে। The Truth about Language: What It Is and Where It Came From বইতে Michael C. Corballis এমনই বিবর্তনীয় যুক্তির অবতারণা করেছেন। আর, এই বইতে পেলাম মানুষের মস্তিষ্ক ও বুদ্ধিমত্তার এমন একটি ব্যাখ্যা যা বিবর্তনবাদের সাথে সাংঘর্ষিক না।
বইটি সুপাঠ্য। নিঃসন্দেহে থট প্রোভোকিং। যা যা চিন্তা এবং সংগ্রহ করে কোহেরেন্ট ন্যারেটিভের রূপ দিয়েছেন, তার অধিকাংশই একাডেমিয়ায় ছড়ানো-ছিটানো আছে। আমাদের মত আমজনতার জন্য তার সবটা সহজলভ্য না। এই দিকটা চিন্তা করলে আসলে প্রচণ্ড দরকারি ছিল বইটি।
ফলাফল: অন ইনটেলিজেন্স।
মানুষের বুদ্ধির (এবং ভাষার) বিকাশ সম্পর্কিত জনপ্রিয় ন্যারেটিভটি কগনিটিভ রেভলিউশনের (Yuval Noah Harari-র কগনিটিভি রেভলিউশন, সোশ্যাল মুভমেন্ট না।) ওপর দাঁড়ানো। পৃথিবীর প্রাণীজগতের সামগ্রিক বিবর্তনের ল্যান্ডস্কেপে এই ন্যারেটিভটি প্রায় মাটি থেকে মানুষ তৈরীর মতই খাপ ছাড়া। সম্প্রতি বুদ্ধিমত্তা ও ভাষার বিকাশের বিবর্তনীয় ব্যাখ্যাগুলো পরিচিতি পাচ্ছে। The Truth about Language: What It Is and Where It Came From বইতে Michael C. Corballis এমনই বিবর্তনীয় যুক্তির অবতারণা করেছেন। আর, এই বইতে পেলাম মানুষের মস্তিষ্ক ও বুদ্ধিমত্তার এমন একটি ব্যাখ্যা যা বিবর্তনবাদের সাথে সাংঘর্ষিক না।
বইটি সুপাঠ্য। নিঃসন্দেহে থট প্রোভোকিং। যা যা চিন্তা এবং সংগ্রহ করে কোহেরেন্ট ন্যারেটিভের রূপ দিয়েছেন, তার অধিকাংশই একাডেমিয়ায় ছড়ানো-ছিটানো আছে। আমাদের মত আমজনতার জন্য তার সবটা সহজলভ্য না। এই দিকটা চিন্তা করলে আসলে প্রচণ্ড দরকারি ছিল বইটি।
November 11, 2011
The book is a take on understanding (human-like) intelligence. The author introduces memory prediction framework to explain the kind of intelligence humans possess. He defines intelligence as the ability to predict. This ability (prediction) can then take different shapes, like decision-making and even creativity. He view the brain as a pattern-recognition device. Different sensory inputs, he claims, are treated in (almost) the same way by the brain.
The treatment of the subject was very pragmatic. His aim to understand the brain was motivated by building intelligent agents that can solve intelligence problems that are (currently) unattainable by machines, e.g. image/sound recognition. This pragmatism is refreshing for those familiar with kurzweil and Hofstadter's view of intelligence (and to some degree consciousness) as they also hint in their work that patterns are a constituent view of information for intelligent machines.
The (almost) identical treatment of patterns is backed up by two main arguments. First, as Ramachandran present in his "A brief tour of human consciousness", different parts of the brain can treat the same sensory input while resulting in an output (e.g. by consciousness) that is consistent with that part; as what happens with people with synesthesia. Second, the structural similarities of the human neocortex. However, this point was particularly surprising since the author, in the first half of the book, attribute intelligence to the neocortex, where it would be expected to be attributed to the cereberal cortex. However, he retain in the second half of the book this conventional view.
In general, the book is a really interesting read. The used language and the progression of ideas are easy to follow. It mixes the right amount of pragmatism and theory for a joyful general read. The only thing that I found bugging was his treatment of the notion of intelligence when he discussed the Turing and Chinese room test. His view that a system is intelligent if it can predict is not really satisfying oppose to the original Turing behavior requirement. Also, I find that his view on the Chinese room test actually contradicts his views in consciousness.
The treatment of the subject was very pragmatic. His aim to understand the brain was motivated by building intelligent agents that can solve intelligence problems that are (currently) unattainable by machines, e.g. image/sound recognition. This pragmatism is refreshing for those familiar with kurzweil and Hofstadter's view of intelligence (and to some degree consciousness) as they also hint in their work that patterns are a constituent view of information for intelligent machines.
The (almost) identical treatment of patterns is backed up by two main arguments. First, as Ramachandran present in his "A brief tour of human consciousness", different parts of the brain can treat the same sensory input while resulting in an output (e.g. by consciousness) that is consistent with that part; as what happens with people with synesthesia. Second, the structural similarities of the human neocortex. However, this point was particularly surprising since the author, in the first half of the book, attribute intelligence to the neocortex, where it would be expected to be attributed to the cereberal cortex. However, he retain in the second half of the book this conventional view.
In general, the book is a really interesting read. The used language and the progression of ideas are easy to follow. It mixes the right amount of pragmatism and theory for a joyful general read. The only thing that I found bugging was his treatment of the notion of intelligence when he discussed the Turing and Chinese room test. His view that a system is intelligent if it can predict is not really satisfying oppose to the original Turing behavior requirement. Also, I find that his view on the Chinese room test actually contradicts his views in consciousness.
May 15, 2016
4.23 stars
Awesome read even if you aren't familiar at all with AI, neural networks, or anything tech related. This book takes a stab at trying to explain how we learn, and breaks down the steps that our brain goes through during the process of learning and recollection.
It's a really good mix of easy to understand, higher level philosophical arguments as well as lots of technical details when he get's into the details about how the neocortex performs it's actions, from sensory input all the way to invariant representations.
Would definitely recommend to anyone who has any interest in understanding how things work (and how our brains understand how things work!).
Awesome read even if you aren't familiar at all with AI, neural networks, or anything tech related. This book takes a stab at trying to explain how we learn, and breaks down the steps that our brain goes through during the process of learning and recollection.
It's a really good mix of easy to understand, higher level philosophical arguments as well as lots of technical details when he get's into the details about how the neocortex performs it's actions, from sensory input all the way to invariant representations.
Would definitely recommend to anyone who has any interest in understanding how things work (and how our brains understand how things work!).
September 29, 2020
I read this book a little late – 15 years to be precise, which is a lifetime when it comes to a dynamic field like Artificial Intelligence and Neuroscience. Therefore, while reading this book, I had to constantly try to go back to 2005 and understand the arguments based on what we knew and believed then. What makes this book remarkable is the author’s view that the only wany to understand intelligence is by studying the only example of it that we know of, our own brain. This may seem obvious to many people, but a large number of Computer Scientists believe that while our brain can provide some clues, it is not essential to understand it in order to create intelligent behavior, just as airplanes are not modeled after birds. The recent commercial successes of neural network based AI systems that are ubiquitous is also another example where the main focus is to get some tasks done rather than asking the question are these systems really intelligent.
The author tried to approach the problem from the brain’s perspective. However, unlike conventional neuroscience, where the focus is mostly empirical, the book tries to create a central theory of the neural organization we see in the newer part of the brain, the neocortex. That is, he tries to find the most fundamental component structure that can then be used to explain all the wide variety of tasks that we can do. So, his basic hypothesis is that the brain is not composed of many specialized parts, but rather it is a huge collection of very similar components, and their organization and interconnection is what accomplishes this magic. He then speculates about what types of circuits can have these properties and tries to find support for these speculations from neurological research.
It is a very compelling set of arguments, and reminded me of the classic of Cybernetics from Ross Ashby – The Design for a Brain. In spite of the age of this book, it remains fresh, thought provoking, and relevant. Since writing this book, Jeff Hawkins created a research organization to push his research forward and refined his theories. In spite of some successes, there are many who consider his work a little too speculative. To me, an outsider, I am glad that there is someone who can take the risk of being speculative. Without a big dose of that, it is unlikely we will crack the problem of intelligence, the mind, and consciousness.
The author tried to approach the problem from the brain’s perspective. However, unlike conventional neuroscience, where the focus is mostly empirical, the book tries to create a central theory of the neural organization we see in the newer part of the brain, the neocortex. That is, he tries to find the most fundamental component structure that can then be used to explain all the wide variety of tasks that we can do. So, his basic hypothesis is that the brain is not composed of many specialized parts, but rather it is a huge collection of very similar components, and their organization and interconnection is what accomplishes this magic. He then speculates about what types of circuits can have these properties and tries to find support for these speculations from neurological research.
It is a very compelling set of arguments, and reminded me of the classic of Cybernetics from Ross Ashby – The Design for a Brain. In spite of the age of this book, it remains fresh, thought provoking, and relevant. Since writing this book, Jeff Hawkins created a research organization to push his research forward and refined his theories. In spite of some successes, there are many who consider his work a little too speculative. To me, an outsider, I am glad that there is someone who can take the risk of being speculative. Without a big dose of that, it is unlikely we will crack the problem of intelligence, the mind, and consciousness.
December 7, 2019
Author is one of the top people in consumer tech (created Palm Pilot), and is deeply interested in AI. He does a pretty good job of presenting a few elements of the field (neural networks, primarily, and that prediction is the most key activity in the neocortex) to a general audience, and then includes some of his own theories and predictions (which is tricky because it's hard for a non-expert to know which parts are broadly accepted and which are his own theories...). Overall, a very interesting book, and since it's nearly 20 years old, it's interesting to see which of his predictions were accurate (things took about 10-20% longer than he predicted, I think, but were much more successful than he predicted); always neat when someone's errors in "wild predictions" are that they were too conservative in some way.
I honestly don't know anywhere near enough about neuroscience to really evaluate that portion or his presentation, but the more general information/cs part was pretty solid.
Most interesting thing to me was the theory that the neocortex evolved to make predictions better in animals, which is a great way to have it merge with the sensory and motor control parts of lower elements of the brain, and provides an incremental and continuous benefit from even slight levels of new capability all the way up to what we have in humans today.
I honestly don't know anywhere near enough about neuroscience to really evaluate that portion or his presentation, but the more general information/cs part was pretty solid.
Most interesting thing to me was the theory that the neocortex evolved to make predictions better in animals, which is a great way to have it merge with the sensory and motor control parts of lower elements of the brain, and provides an incremental and continuous benefit from even slight levels of new capability all the way up to what we have in humans today.
January 24, 2019
To be honest, I have very mixed feelings about this book. I really enjoyed the first part, it gave a good introduction and discussion of neuroscientific research. Even though I didn't agree on most points (being somewhat a neuroscientist myself), I think it was well written and ~interesting~. But, I am not quite sure how to feel about the whole second part: his theory. Hawkins surely has good points, but I'm missing some kind of evidence for his ideas. Overall, I was a bit disappointed by the lack of references: even if you write a popscience book, it's nice to actually cite the studies you're talking about - especially so the interested reader can learn more. I also found there was too much of Hawkins person in the book and too little of the actual content. I had hoped for a book that would give a nice sort of introduction into the field, not explain a new theory without evidence for it. Overall I felt a lot of it was too subjective, too "I founded this company, and made this important invention", especially since his inventions actually ended up being pretty not important 15 years later.
Nonetheless, an easy-ish read with loads of good parts and ideas. Definitely suitable for people without any prior knowledge but should be taken with a big, BIG grain of salt.
Nonetheless, an easy-ish read with loads of good parts and ideas. Definitely suitable for people without any prior knowledge but should be taken with a big, BIG grain of salt.
April 15, 2020
An amazing book with an amazing framework of how our brains work . It was a fantastic read . This book lays out the framework for understanding brain . I think this is the most important book in neuroscience , psycology and artificial intelligence for sure . This book is the must read for everybody who is curious about brain and how it works . Sections of creativity , intelligence , power of imagination , psycology are very thoughtful and charming . Jeff hawkins will definitely change the way you think about intelligence .
Worth my time .
Worth my time .
February 21, 2017
The book gets 5 stars for having been written in 2002 and only just now coming to the point at which the future has surpassed some minor aspects of what he's saying.
I really liked the way he addresses the topic of intelligence and what it really means to think about it in a non-linear way. If your going down this track of thought, On Intelligence sticks out.
P. 51 For example, a framework of mechanistic development of systems where the location is the difference vs. modularizing in a way that does not really understand the flexibility created by nature via the brain. Brilliant if you're thinking at all of building systems.
P. 69 The idea of the forgotten time element and what that means to visual and auditory inputs via the brain vs. what we are programming. Very insightful.
P. 81 The way your brain differentiates important vs. unimportant in storage into long term memory (for example, memorizing a song by tonal modulation vs. perfect pitch) and the ability to recall even if the pitch differs. Really great observation.
Lots of great stuff on pattern recognition and how to think of the brain as a pattern recognition machine.
Some of his thoughts on speech recognition has come a long way as with a few other ways that people now think of AI.
Still, great book for a beginner like me.
I really liked the way he addresses the topic of intelligence and what it really means to think about it in a non-linear way. If your going down this track of thought, On Intelligence sticks out.
P. 51 For example, a framework of mechanistic development of systems where the location is the difference vs. modularizing in a way that does not really understand the flexibility created by nature via the brain. Brilliant if you're thinking at all of building systems.
P. 69 The idea of the forgotten time element and what that means to visual and auditory inputs via the brain vs. what we are programming. Very insightful.
P. 81 The way your brain differentiates important vs. unimportant in storage into long term memory (for example, memorizing a song by tonal modulation vs. perfect pitch) and the ability to recall even if the pitch differs. Really great observation.
Lots of great stuff on pattern recognition and how to think of the brain as a pattern recognition machine.
Some of his thoughts on speech recognition has come a long way as with a few other ways that people now think of AI.
Still, great book for a beginner like me.
September 25, 2013
Enjoyed thoroughly. Read it twice.
May 25, 2020
we humans are just a brain and a body. brain in this equation is the integral part, it gives us the identity and makes us what we are. you lose a limb, you just lose a limp; you lose a brain, you lose everything.
it is as complicated an organ as important it is, but science still doesnot know how it works, we just know the basics of how neurons interact, how signals are passed and which area is doing what work.
jeff hawkins, a neuroscientist, is trying to propose his theory of how, he thinks, brain works. it is just a theory which does not completely mean that it is exactly how the brain works but it surely is a step towards that quest.
this book will provoke your thoughts and stimulate more questions. i am putting some ideas presented in this book here:
whatever we are seeing is portrayed on retina of our eye, the brain doesn’t see the image. All our senses are reporting their inputs to the brain and it is producing an output. All the inputs (from eyes, ears, etc) are converted to the same thing and our brain follows one universal algorithm for all of them. Pointing to the thought that the kind of input doesnot matter, it is how it is perceived in the brain. because it is following mostly one single kind of algorithm for all the work. Other senses can be trained to cope up for the missing senses, and may be that’s why blind people are good at hearing.
it talks about memory prediction framework, which says all of our experiences are being stored in our memory(how and where in the book) and then we predict everything based on that memory, and then behave accordingly. prediction is the key essence of human intelligence because every time we are predicting, if things go according to our predictions we call them normal and when they don’t we feel absurd. we can walk in our rooms without light because the structure of the room is in our memory and we can predict well which thing is where.
it says, we all make a model of this world from our memory, experiences, and surroundings. We predict things and make decisions based on that model. most of things in our models are same but not everything. no one has a perfect model in their brains. what we call reality is based on that model. that’s why people give different weights to things because in one person’s model that thing is important but not in other’s model. This way the early years of someone’s life are the most important ones.
if we completely understand how brain works, we will be able to replicate it and humanity will see new horizons.
this is a review of this book, not my thoughts. my quest for how brain works will continue.
it is as complicated an organ as important it is, but science still doesnot know how it works, we just know the basics of how neurons interact, how signals are passed and which area is doing what work.
jeff hawkins, a neuroscientist, is trying to propose his theory of how, he thinks, brain works. it is just a theory which does not completely mean that it is exactly how the brain works but it surely is a step towards that quest.
this book will provoke your thoughts and stimulate more questions. i am putting some ideas presented in this book here:
whatever we are seeing is portrayed on retina of our eye, the brain doesn’t see the image. All our senses are reporting their inputs to the brain and it is producing an output. All the inputs (from eyes, ears, etc) are converted to the same thing and our brain follows one universal algorithm for all of them. Pointing to the thought that the kind of input doesnot matter, it is how it is perceived in the brain. because it is following mostly one single kind of algorithm for all the work. Other senses can be trained to cope up for the missing senses, and may be that’s why blind people are good at hearing.
it talks about memory prediction framework, which says all of our experiences are being stored in our memory(how and where in the book) and then we predict everything based on that memory, and then behave accordingly. prediction is the key essence of human intelligence because every time we are predicting, if things go according to our predictions we call them normal and when they don’t we feel absurd. we can walk in our rooms without light because the structure of the room is in our memory and we can predict well which thing is where.
it says, we all make a model of this world from our memory, experiences, and surroundings. We predict things and make decisions based on that model. most of things in our models are same but not everything. no one has a perfect model in their brains. what we call reality is based on that model. that’s why people give different weights to things because in one person’s model that thing is important but not in other’s model. This way the early years of someone’s life are the most important ones.
if we completely understand how brain works, we will be able to replicate it and humanity will see new horizons.
this is a review of this book, not my thoughts. my quest for how brain works will continue.
August 6, 2019
The brain as a "pattern device" that works through memory
"Prediction is not just one of the things your brain does. It is the primary function of the neocortex, and the foundation of intelligence." (p. 89)
Perhaps the crux of Hawkins's insight into how our brains work and how that is different from how computers work can be gleaned from considering how to catch a ball in flight.
It used to be thought that such tasks were solved by the brain through calculation. The brain would calculate the flight of the ball, adjusting the muscles of the body appropriately so as to arrive at a spot where the ball would be and grab it. Artificial intelligence people working on robots used this method and found out that it was enormously complex, so much so that the robots remained clumsy (and not about to play centerfield for the New York Yankees).
What Hawkins is saying is that the brain does NOT calculate the flight of the ball but instead recalls from memory similar flights of balls while at the same time recalling again from memory the muscular workings of the body as it went after and caught or did not catch similar balls in flight. After a bit of practice (storing memories) a person can get very good at catching balls.
In other words the brain predicts where the ball is going to be not through a laborious and lengthy calculation but through memories of similar events. This is a startling insight. Hawkins shows how everything we do is based on our brain's ability to predict events based on previous experience. Here's how it works:
First there is a "sequence of patterns" of past events stored in the brain.
Second, the brain has an "auto-associative mechanism" that allows it to "recall complete patterns when given only partial or distorted inputs." (p. 73) Unlike computer intelligence, human intelligence can figure out that "Wass up?" means the same thing as "What's up?" or that a face seen from one angle is the same as that face seen from another angle or even seen in some sort of distortion. This is something computers cannot reliably do.
Third, the brain stores "invariant representations" of things seen, heard, felt, etc. "Invariant" in this context means unaffected by differences in light or tone or inflection or background or any one of millions of small, inessential differences that could throw us off. These representations are not exact. They are in a way like Plato's ideal forms except they are not ideal but generalized. They are memories of the relationships between and among various features. In the case of a human face, Hawkins writes that what makes a face recognizable "are its relative dimensions, relative colors, and relative proportions, not how it appeared one instant last Tuesday at lunch." (p. 81)
Hawkins's definition of intelligence in terms of predictive ability is what I found most exciting in the book. When people talk about intelligence I usually want to demand "intelligence for what?" since the criteria for defining intelligence has always been so muddied. One of the ways of establishing a theory in science is through its ability to make accurate predictions. To judge the brain the same way seems strikingly right. Not only that but no longer do we have to beg the question of what intelligence is. It is the ability to predict.
These predictions are about everything in our lives and they involve all of our senses. As Hawkins puts it, "All regions of your neocortex are simultaneously trying to predict what their next experience will be. Visual areas make predictions about edges, shapes, objects, locations, and motions. Auditory areas make predictions about tones, direction to source, and patterns of sound. Somatosensory areas make predictions about touch, texture, contour, and temperature." (pp. 88-89)
While the first five chapters are eminently readable and exciting, Chapter 6, "How the Cortex Works" (the longest in the book) might be a bit tedious and technical for the general reader. (I know it was for me.)
In Chapter 7, "Consciousness and Creativity" Hawkins writes, "Most of what you perceive is not coming through your senses; it is generated by your internal memory model." (p. 202) We do not experience the world directly and we do not interpret it objectively. Our predictions in a sense are prejudices or stereotypes that sometimes lead us astray. Hawkins writes, "…you could substitute the word 'stereotype' for 'invariant memory'…without substantially altering the meaning. Prediction by analogy is pretty much the same as judgment by stereotype." (p. 203)
In the final chapter, "The Future of Intelligence" Hawkins makes it clear that intelligent machines will not be taking over the world. He writes, "The computer in your home, or the Internet, has as much chance of spontaneously turning sentient as does a cash register." (p. 214) Furthermore, an intelligent machine "will not have a mind that is remotely humanlike unless we imbue it with humanlike emotional systems and humanlike experiences. That would be extremely difficult and, it seems to me, quite pointless." (p. 208). Finally, fears that machines will take over the world "rest on a false analogy…a conflation of intelligence…with the emotional drives of the old brain--things like fear, paranoia, and desire. But intelligent machines will not have these faculties. They will not have personal ambition. They will not desire wealth, social recognition, or sensual gratification. They will not have appetites, addictions, or mood disorders." (p. 216)
Hawkins goes on to predict that, with an approach based on learning and memory instead of brute calculation, we will build truly intelligent machines, the applications of which will be numerous and include applications impossible to predict.
I would like to point out that Hawkins' idea that our cortex is continually making predictions about the environment, predictions that we scarcely notice unless they are wrong, is similar to an idea that John McCrone presented in his book Going Inside: A Tour Round a Single Moment of Consciousness (2001), a book I also highly recommend.
--Dennis Littrell, author of “The World Is Not as We Think It Is”
"Prediction is not just one of the things your brain does. It is the primary function of the neocortex, and the foundation of intelligence." (p. 89)
Perhaps the crux of Hawkins's insight into how our brains work and how that is different from how computers work can be gleaned from considering how to catch a ball in flight.
It used to be thought that such tasks were solved by the brain through calculation. The brain would calculate the flight of the ball, adjusting the muscles of the body appropriately so as to arrive at a spot where the ball would be and grab it. Artificial intelligence people working on robots used this method and found out that it was enormously complex, so much so that the robots remained clumsy (and not about to play centerfield for the New York Yankees).
What Hawkins is saying is that the brain does NOT calculate the flight of the ball but instead recalls from memory similar flights of balls while at the same time recalling again from memory the muscular workings of the body as it went after and caught or did not catch similar balls in flight. After a bit of practice (storing memories) a person can get very good at catching balls.
In other words the brain predicts where the ball is going to be not through a laborious and lengthy calculation but through memories of similar events. This is a startling insight. Hawkins shows how everything we do is based on our brain's ability to predict events based on previous experience. Here's how it works:
First there is a "sequence of patterns" of past events stored in the brain.
Second, the brain has an "auto-associative mechanism" that allows it to "recall complete patterns when given only partial or distorted inputs." (p. 73) Unlike computer intelligence, human intelligence can figure out that "Wass up?" means the same thing as "What's up?" or that a face seen from one angle is the same as that face seen from another angle or even seen in some sort of distortion. This is something computers cannot reliably do.
Third, the brain stores "invariant representations" of things seen, heard, felt, etc. "Invariant" in this context means unaffected by differences in light or tone or inflection or background or any one of millions of small, inessential differences that could throw us off. These representations are not exact. They are in a way like Plato's ideal forms except they are not ideal but generalized. They are memories of the relationships between and among various features. In the case of a human face, Hawkins writes that what makes a face recognizable "are its relative dimensions, relative colors, and relative proportions, not how it appeared one instant last Tuesday at lunch." (p. 81)
Hawkins's definition of intelligence in terms of predictive ability is what I found most exciting in the book. When people talk about intelligence I usually want to demand "intelligence for what?" since the criteria for defining intelligence has always been so muddied. One of the ways of establishing a theory in science is through its ability to make accurate predictions. To judge the brain the same way seems strikingly right. Not only that but no longer do we have to beg the question of what intelligence is. It is the ability to predict.
These predictions are about everything in our lives and they involve all of our senses. As Hawkins puts it, "All regions of your neocortex are simultaneously trying to predict what their next experience will be. Visual areas make predictions about edges, shapes, objects, locations, and motions. Auditory areas make predictions about tones, direction to source, and patterns of sound. Somatosensory areas make predictions about touch, texture, contour, and temperature." (pp. 88-89)
While the first five chapters are eminently readable and exciting, Chapter 6, "How the Cortex Works" (the longest in the book) might be a bit tedious and technical for the general reader. (I know it was for me.)
In Chapter 7, "Consciousness and Creativity" Hawkins writes, "Most of what you perceive is not coming through your senses; it is generated by your internal memory model." (p. 202) We do not experience the world directly and we do not interpret it objectively. Our predictions in a sense are prejudices or stereotypes that sometimes lead us astray. Hawkins writes, "…you could substitute the word 'stereotype' for 'invariant memory'…without substantially altering the meaning. Prediction by analogy is pretty much the same as judgment by stereotype." (p. 203)
In the final chapter, "The Future of Intelligence" Hawkins makes it clear that intelligent machines will not be taking over the world. He writes, "The computer in your home, or the Internet, has as much chance of spontaneously turning sentient as does a cash register." (p. 214) Furthermore, an intelligent machine "will not have a mind that is remotely humanlike unless we imbue it with humanlike emotional systems and humanlike experiences. That would be extremely difficult and, it seems to me, quite pointless." (p. 208). Finally, fears that machines will take over the world "rest on a false analogy…a conflation of intelligence…with the emotional drives of the old brain--things like fear, paranoia, and desire. But intelligent machines will not have these faculties. They will not have personal ambition. They will not desire wealth, social recognition, or sensual gratification. They will not have appetites, addictions, or mood disorders." (p. 216)
Hawkins goes on to predict that, with an approach based on learning and memory instead of brute calculation, we will build truly intelligent machines, the applications of which will be numerous and include applications impossible to predict.
I would like to point out that Hawkins' idea that our cortex is continually making predictions about the environment, predictions that we scarcely notice unless they are wrong, is similar to an idea that John McCrone presented in his book Going Inside: A Tour Round a Single Moment of Consciousness (2001), a book I also highly recommend.
--Dennis Littrell, author of “The World Is Not as We Think It Is”
October 18, 2020
Notes:
1. Brain processes data from different senses (touch, smell, vision, etc.) similarly - it tells the patterns from the input. From an aesthetic perspective, this makes a lot of sense. I remember seeing in literature the examples of using one sense to describe another. And there's Kandinsky who paints music.
2. Memory. It mentions a lot of memories are not forgotten, but just stay somewhere. With triggers, we recall things that we haven't think of for a long time. It is fascinating that in the pandemic, I've picked up so many memory gems that had been collecting dust in the dark corners.
1. Brain processes data from different senses (touch, smell, vision, etc.) similarly - it tells the patterns from the input. From an aesthetic perspective, this makes a lot of sense. I remember seeing in literature the examples of using one sense to describe another. And there's Kandinsky who paints music.
2. Memory. It mentions a lot of memories are not forgotten, but just stay somewhere. With triggers, we recall things that we haven't think of for a long time. It is fascinating that in the pandemic, I've picked up so many memory gems that had been collecting dust in the dark corners.
March 27, 2012
Mr Hawkins' dream was to encapsulate a basic theory of intelligence in a straightforward plainly written book. Written with science writer Sandra Blakeslee, "On Intelligence" combines Mr Hawkins' motivational autobiography, a scientific treatise on natural and artificial intelligence, and a philosophical discussion delivered in a no-nonsense, unembellished, yet stimulating narrative.
At its core, "On Intelligence" postulates that all higher cognitive functions are built on a single relatively simple algorithm replicated across the neocortex. This hypothetic "basic cortical algorithm" is described as a predictive autoassociative hierarchical network. Left to its own devices, such a neural network should spontaneously generate stable invariant representations of regularities in the environment giving birth to perception, behavior, thoughts, consciousness, and imagination. If we could only mimic Nature and build such a network in silicon, we should be able to make computers that learn, think, and imagine. Mr Hawkins admits that most of these ideas are not original and his contribution is to organize them into a coherent hypothetical framework.
How credible is Mr Hawkins' hypothesis? How do we know the brain does this? How do we know that such an artificial model would exhibit animal-like intelligence? Mr Hawkins' answer is: be optimistic -- we are way overdue for some kind of a general theory of the brain. In a break from scientific form, Mr Hawkins does not seek out contradictory evidence. The autobiographical sections carries an air of a quixotic struggle against the errors and prejudices of the scientific and corporate establishments of the past and present, who lack the audacity to imagine that a comprehensive theory of intelligence could be within reach. In its more technical sections, the book identifies specific cortical structures responsible for these computations in rather computational than biological terms. No experimental evidence and no working computer models are described or reviewed critically. Instead, the key premises derive from introspection and personal interviews with authorities on the subject, e.g. "I had spoken to several ... experts and asked them to explain..." Mr Hawkins mixes experimentally supported findings with speculation and swiftly decides standing controversies without identifying them as such, leaving a casual reader with an exaggerated impression of how much is understood about cognition. In this way, the book often reads rather like marketing material for a specific approach than a thoroughly researched thesis presenting latest scientific findings.
Every neuroscientist strives to intuit a fundamental principle behind the ocean of facts about the nervous system and every computer scientists dreams of creating systems that could develop intelligence. Yet Nature is slow to give up her recipes. By helping envision what the answers could be, "On Intelligence" stands to inspire the budding scientist and engineer with the confidence to probe into the most daunting natural phenomenon that is intelligence. And it is for its enthusiasm and inspiration that "On Intelligence" earns my four stars.
At its core, "On Intelligence" postulates that all higher cognitive functions are built on a single relatively simple algorithm replicated across the neocortex. This hypothetic "basic cortical algorithm" is described as a predictive autoassociative hierarchical network. Left to its own devices, such a neural network should spontaneously generate stable invariant representations of regularities in the environment giving birth to perception, behavior, thoughts, consciousness, and imagination. If we could only mimic Nature and build such a network in silicon, we should be able to make computers that learn, think, and imagine. Mr Hawkins admits that most of these ideas are not original and his contribution is to organize them into a coherent hypothetical framework.
How credible is Mr Hawkins' hypothesis? How do we know the brain does this? How do we know that such an artificial model would exhibit animal-like intelligence? Mr Hawkins' answer is: be optimistic -- we are way overdue for some kind of a general theory of the brain. In a break from scientific form, Mr Hawkins does not seek out contradictory evidence. The autobiographical sections carries an air of a quixotic struggle against the errors and prejudices of the scientific and corporate establishments of the past and present, who lack the audacity to imagine that a comprehensive theory of intelligence could be within reach. In its more technical sections, the book identifies specific cortical structures responsible for these computations in rather computational than biological terms. No experimental evidence and no working computer models are described or reviewed critically. Instead, the key premises derive from introspection and personal interviews with authorities on the subject, e.g. "I had spoken to several ... experts and asked them to explain..." Mr Hawkins mixes experimentally supported findings with speculation and swiftly decides standing controversies without identifying them as such, leaving a casual reader with an exaggerated impression of how much is understood about cognition. In this way, the book often reads rather like marketing material for a specific approach than a thoroughly researched thesis presenting latest scientific findings.
Every neuroscientist strives to intuit a fundamental principle behind the ocean of facts about the nervous system and every computer scientists dreams of creating systems that could develop intelligence. Yet Nature is slow to give up her recipes. By helping envision what the answers could be, "On Intelligence" stands to inspire the budding scientist and engineer with the confidence to probe into the most daunting natural phenomenon that is intelligence. And it is for its enthusiasm and inspiration that "On Intelligence" earns my four stars.
February 27, 2016
If to put it short, Palm Pilot-inventor Jeff Hawkins book explains his memory-prediction framework theory of the brain and describes some of its consequences.
Well, that's makes sense to me, as I've learned from professor Wang in my first lectures of neuroscience: "Brain is just a surviving organ...".
Sure, but there is always the Homunculus, that little bastard that is preventing us to perceive our realm directly, straightforwardly. Every time we turn to study ourselves we get into mess with that little guy, who interferes with us,...with us, with the real owner of our body and we found ourselves trapped with lost framework, with no philosophical background that would support our theories.
To comprehend Mr. Hawkins theory, the reader must accept what Mr. Thomas Metzinger believes strongly that it is possible to solve the philosophical puzzle of consciousness only if we come to understand that to the best of our current knowledge there is no thing, no indivisible entity that is us, neither in the brain nor in some metaphysical realm beyond this world.
So, to enjoy and understand this book, start with eliminating the "Homunculus", accept that the voice and dialog occurring in our heads is normal communication between brain modules that competes for the action of the body to keep us alive, and join Mr. Hawkins in trip into unknown of neurobiology and neuroscience generally.
To comprehend intelligence and AI, one must acquire proper framework, and learn about the brain and entire nervous system thoroughly. From point of view of AI however, one vital part of the brain is essential to fully comprehend- the Cortex! This is the trip that will lead you deep into the 6-layered magic cover that is making us exactly what we are. Have a nice trip!
Well, that's makes sense to me, as I've learned from professor Wang in my first lectures of neuroscience: "Brain is just a surviving organ...".
Sure, but there is always the Homunculus, that little bastard that is preventing us to perceive our realm directly, straightforwardly. Every time we turn to study ourselves we get into mess with that little guy, who interferes with us,...with us, with the real owner of our body and we found ourselves trapped with lost framework, with no philosophical background that would support our theories.
To comprehend Mr. Hawkins theory, the reader must accept what Mr. Thomas Metzinger believes strongly that it is possible to solve the philosophical puzzle of consciousness only if we come to understand that to the best of our current knowledge there is no thing, no indivisible entity that is us, neither in the brain nor in some metaphysical realm beyond this world.
So, to enjoy and understand this book, start with eliminating the "Homunculus", accept that the voice and dialog occurring in our heads is normal communication between brain modules that competes for the action of the body to keep us alive, and join Mr. Hawkins in trip into unknown of neurobiology and neuroscience generally.
To comprehend intelligence and AI, one must acquire proper framework, and learn about the brain and entire nervous system thoroughly. From point of view of AI however, one vital part of the brain is essential to fully comprehend- the Cortex! This is the trip that will lead you deep into the 6-layered magic cover that is making us exactly what we are. Have a nice trip!
August 27, 2007
original review:
http://www.reemer.com/archives/2005/0...
This is the second book that Phil Terry asked us to read as part of the Creative Good fellows program. It was writted by Jeff Hawkins, creator of the PalmPilot and Treo. Turns out Jeff's other passion is trying to understand how the brain works.
This book lays out his theory of how the mind works in layman's terms. Hawkins premise is that the brain uses a "memory-prediction" framework to operate, and states that his model fills in a lot of holes in existing models of how the brain works. In other words, the brain store past experiences as patterns, and uses those patterns to predict how future events will occur. If future events differ from the predictions, the brain parses the new patterns and adjusts its predictions accordingly.
The model is simple at heart, and makes a lot of sense to this layman. Hawkins' goal is to ultimately build computerized brains, and expresses disappointment at current efforts to do so. This book is a well thought-out argument that thumbs its nose at current thinking, and in my experience, this will only help in achieving Hawkins' ultimate goal of building a functioning brain.
http://www.reemer.com/archives/2005/0...
This is the second book that Phil Terry asked us to read as part of the Creative Good fellows program. It was writted by Jeff Hawkins, creator of the PalmPilot and Treo. Turns out Jeff's other passion is trying to understand how the brain works.
This book lays out his theory of how the mind works in layman's terms. Hawkins premise is that the brain uses a "memory-prediction" framework to operate, and states that his model fills in a lot of holes in existing models of how the brain works. In other words, the brain store past experiences as patterns, and uses those patterns to predict how future events will occur. If future events differ from the predictions, the brain parses the new patterns and adjusts its predictions accordingly.
The model is simple at heart, and makes a lot of sense to this layman. Hawkins' goal is to ultimately build computerized brains, and expresses disappointment at current efforts to do so. This book is a well thought-out argument that thumbs its nose at current thinking, and in my experience, this will only help in achieving Hawkins' ultimate goal of building a functioning brain.
December 25, 2013
هذا الكتاب .. و حياتي يدوران حول شغفين
الاول هو برمجة الكمبيوترات المحمولة و الثاني هو الذكاء
هكذا بدأ المؤلف الكتاب
لا اعتقد اني ابالغ ... الكتاب ثورة في العلم و يؤدي حاليا الي ثورة في التقنية موضوع الكتاب ببساطة شديدة هو ان الذكاء البشري هو نوع من انواع المعالجة التي يمكن محاكاتها باستخدام الحاسب
الفكرة هنا مختلفة تماما عن علم (الشبكات العصبية) فهي تقترب من فكرة (كيف يعمل المخ البشري) اعتقد ان الكثير من الابحاث في الادراك و علم الخلايا العصبية و علم النفس قد وصلت الي نتائج حتي الان تؤكد صحة النظرية
يحلم الكاتب بصنع الة يستخدم في بنيتها نظاما للذكاء يشبه الذكاء البشري
هل ممكن؟
الاول هو برمجة الكمبيوترات المحمولة و الثاني هو الذكاء
هكذا بدأ المؤلف الكتاب
لا اعتقد اني ابالغ ... الكتاب ثورة في العلم و يؤدي حاليا الي ثورة في التقنية موضوع الكتاب ببساطة شديدة هو ان الذكاء البشري هو نوع من انواع المعالجة التي يمكن محاكاتها باستخدام الحاسب
الفكرة هنا مختلفة تماما عن علم (الشبكات العصبية) فهي تقترب من فكرة (كيف يعمل المخ البشري) اعتقد ان الكثير من الابحاث في الادراك و علم الخلايا العصبية و علم النفس قد وصلت الي نتائج حتي الان تؤكد صحة النظرية
يحلم الكاتب بصنع الة يستخدم في بنيتها نظاما للذكاء يشبه الذكاء البشري
هل ممكن؟
March 17, 2011
Interesting high-level theory of how the neocortex works, and a call to create "intelligent" machines that use the same algorithm/structure to perform pattern matching, hierarchical learning and prediction
October 11, 2013
I can't really say this was a practical book but it definitely gives a different perspective on how the brain works and how the current AI implementations are totally off the target. It's enlightening. Worth the read if you are a software developer for sure.
July 4, 2016
Key for understanding the different abstraction layers of how likely our brain and intelligence works. Key for understanding new developments in artificial neural networks too
February 17, 2014
This engaging, non(too)technical book offers a new and plausible theory of how the brain, or more specifically the neocortex, works.
When I learned about the existence of this book, I was drawn to it for a number of reasons. For one thing, I'm intrigued by the faculty we call intelligence: what is it, exactly? For another, I, like the author Jeff Hawkins, have long been fascinated by the brain and how it works. And finally I was eager to read a book on neuroscience by a nonscientist, for Hawkins, inventor of the Palm Pilot and other things, is a technologist who has long pursued brain science as a hobby. I love the idea of contributions to knowledge being made by amateurs, for they seem best able to think outside the box.
And thinking outside the box is what Hawkins has done here. His point of attack was to discover whether it is possible to build an "intelligent machine," and how this might be done. He noted the relative unsatisfactoriness of the results achieved by "artificial intelligence" in the computing world, and wondered why this was. How was it that a computer, with processors executing millions of instructions per second, could not seem to remotely approach the prowess of the human brain at most tasks requiring "intelligence," when the cells in that brain could only execute a few tens of "instructions" per second? Even relatively simple perceptual tasks, like recognizing faces and chairs, are done effortlessly and almost instantly by humans, while machines toil to achieve a success rate well below 100%. What have humans got that computers don't got?
Humans have got a way of processing information that is completely different from the way computers process it. The brain, unlike a computer, does not run on the instructions of a single master program controlled from the top. The brain, says Hawkins, operates as a vast array of small, localized processing systems. In particular, the neocortex--that sheet of neurons that covers the upper frontal part of our brain, and is responsible for all of our human intelligence--is set up as an intricate, interconnected feedback system that can be boiled down to performing two functions: memory and prediction. He's saying that what we call intelligence is the interplay of memory and prediction.
To defend and illustrate this thesis, he goes into some detail on exactly how he thinks the neocortex is wired up. It is known to consist of 6 layers of neurons, which are all interconnected in certain characteristic ways. Hawkins shows why they are so interconnected, and how this results in the formation of memories at increasingly high levels of abstraction. What we call intelligence is the recognition of a current sensory input as belonging to an abstract category already in memory. According to this view, animals that also possess neocortexes have this same intelligence, but in lower degree than humans, who have the most sophisticated neocortex (one interesting fact in the book was that dolphins, which are intelligent and also possess large brains, have a neocortex with only 3 layers, as opposed to the humans' 6).
Hawkins makes his case very well; I found it persuasive. Where I found myself less persuaded was in what I would call the philosophical side of the book, where the author addresses questions such as, What is creativity? What is imagination? What is consciousness? And, of course, the basic question: what is intelligence itself? I think that Hawkins, an extremely able technologist and even scientist, overplays his hand as a philosopher.
Along the way, for example, he talks about Plato's theory of forms as an explanation of how we are able to recognize sameness in the hurly-burly of our ever-shifting sensations. Hawkins notes offhand, "His system of explanation was wildly off the mark." Well, maybe it was and maybe it wasn't, but what makes Hawkins so sure? I recall that Roger Penrose, in chapter 1 of his Road To Reality, treats the world of mathematical truth as one part of a Platonic world of forms, seemingly real but also different from the worlds of mental experience and of physical things. My point here is just that Plato's ideas live on; they'll keep climbing back out of the dustbin of history.
I had similar feelings about Hawkins's take on the other philosophical questions. He contends that there the difference between the intelligence of, say, a rat and of a human is purely one of degree. But Mortimer J. Adler, in his book Intellect: Mind over Matter, contends the opposite. According to him, intellect--which was the word formerly used to label the faculties that we now point to with the word intelligence--is something more than the rudimentary power of abstraction used by brutes. In this view, animals are able to respond to individually differing things in the same way, as when a rat is able to press different triangular buttons to get food, but this is not the same thing as
Adler argues that the brain is a necessary but not a sufficient condition of the human intellect. The existence of the intellect, he thinks, is a sign that a human being is something more than just a body.
Is Adler right in this? I don't know. And I don't think Jeff Hawkins knows either, no matter how confident he is in his assertions. But for someone who wants to build intelligent machines, I think a cautious outlook would be fitting. Hawkins dismisses people's worries that superintelligent machines might become our overlords or our executioners, like HAL in 2001: A Space Odyssey or the Skynet computers in the Terminator movies. He thinks that such behavior would require the presence of the equivalent of the emotional centers of the brain in addition to the neocortex, and he's only planning to build an analog of the neocortex. So don't worry, folks.
I recall reading a comment from the Dalai Lama, apparently changing his mind about whether robots or machines could become sentient. He said that if some being had the necessary karma to take birth or manifestation in a machine, then that would happen. I note that karma is not a word that occurs in the index of Hawkins' book.
But this is a good, clear, strongly argued, plainspoken, provocative, and, yes, intelligent book. Hawkins has persuaded me that "intelligent" machines are very likely in our near future. And I'm sure they will be very helpful and will have no reason to do anything mean to us, their intellectual inferiors.
If they do turn on us, then, well, maybe God will help us.
When I learned about the existence of this book, I was drawn to it for a number of reasons. For one thing, I'm intrigued by the faculty we call intelligence: what is it, exactly? For another, I, like the author Jeff Hawkins, have long been fascinated by the brain and how it works. And finally I was eager to read a book on neuroscience by a nonscientist, for Hawkins, inventor of the Palm Pilot and other things, is a technologist who has long pursued brain science as a hobby. I love the idea of contributions to knowledge being made by amateurs, for they seem best able to think outside the box.
And thinking outside the box is what Hawkins has done here. His point of attack was to discover whether it is possible to build an "intelligent machine," and how this might be done. He noted the relative unsatisfactoriness of the results achieved by "artificial intelligence" in the computing world, and wondered why this was. How was it that a computer, with processors executing millions of instructions per second, could not seem to remotely approach the prowess of the human brain at most tasks requiring "intelligence," when the cells in that brain could only execute a few tens of "instructions" per second? Even relatively simple perceptual tasks, like recognizing faces and chairs, are done effortlessly and almost instantly by humans, while machines toil to achieve a success rate well below 100%. What have humans got that computers don't got?
Humans have got a way of processing information that is completely different from the way computers process it. The brain, unlike a computer, does not run on the instructions of a single master program controlled from the top. The brain, says Hawkins, operates as a vast array of small, localized processing systems. In particular, the neocortex--that sheet of neurons that covers the upper frontal part of our brain, and is responsible for all of our human intelligence--is set up as an intricate, interconnected feedback system that can be boiled down to performing two functions: memory and prediction. He's saying that what we call intelligence is the interplay of memory and prediction.
To defend and illustrate this thesis, he goes into some detail on exactly how he thinks the neocortex is wired up. It is known to consist of 6 layers of neurons, which are all interconnected in certain characteristic ways. Hawkins shows why they are so interconnected, and how this results in the formation of memories at increasingly high levels of abstraction. What we call intelligence is the recognition of a current sensory input as belonging to an abstract category already in memory. According to this view, animals that also possess neocortexes have this same intelligence, but in lower degree than humans, who have the most sophisticated neocortex (one interesting fact in the book was that dolphins, which are intelligent and also possess large brains, have a neocortex with only 3 layers, as opposed to the humans' 6).
Hawkins makes his case very well; I found it persuasive. Where I found myself less persuaded was in what I would call the philosophical side of the book, where the author addresses questions such as, What is creativity? What is imagination? What is consciousness? And, of course, the basic question: what is intelligence itself? I think that Hawkins, an extremely able technologist and even scientist, overplays his hand as a philosopher.
Along the way, for example, he talks about Plato's theory of forms as an explanation of how we are able to recognize sameness in the hurly-burly of our ever-shifting sensations. Hawkins notes offhand, "His system of explanation was wildly off the mark." Well, maybe it was and maybe it wasn't, but what makes Hawkins so sure? I recall that Roger Penrose, in chapter 1 of his Road To Reality, treats the world of mathematical truth as one part of a Platonic world of forms, seemingly real but also different from the worlds of mental experience and of physical things. My point here is just that Plato's ideas live on; they'll keep climbing back out of the dustbin of history.
I had similar feelings about Hawkins's take on the other philosophical questions. He contends that there the difference between the intelligence of, say, a rat and of a human is purely one of degree. But Mortimer J. Adler, in his book Intellect: Mind over Matter, contends the opposite. According to him, intellect--which was the word formerly used to label the faculties that we now point to with the word intelligence--is something more than the rudimentary power of abstraction used by brutes. In this view, animals are able to respond to individually differing things in the same way, as when a rat is able to press different triangular buttons to get food, but this is not the same thing as
cognizing what is common to them or knowing them in their universal aspects. . . . By means of concepts, and only by means of concepts, we understand kinds or classes as such entirely apart from perceived particulars and even though no particular instances exist.
Adler argues that the brain is a necessary but not a sufficient condition of the human intellect. The existence of the intellect, he thinks, is a sign that a human being is something more than just a body.
Is Adler right in this? I don't know. And I don't think Jeff Hawkins knows either, no matter how confident he is in his assertions. But for someone who wants to build intelligent machines, I think a cautious outlook would be fitting. Hawkins dismisses people's worries that superintelligent machines might become our overlords or our executioners, like HAL in 2001: A Space Odyssey or the Skynet computers in the Terminator movies. He thinks that such behavior would require the presence of the equivalent of the emotional centers of the brain in addition to the neocortex, and he's only planning to build an analog of the neocortex. So don't worry, folks.
I recall reading a comment from the Dalai Lama, apparently changing his mind about whether robots or machines could become sentient. He said that if some being had the necessary karma to take birth or manifestation in a machine, then that would happen. I note that karma is not a word that occurs in the index of Hawkins' book.
But this is a good, clear, strongly argued, plainspoken, provocative, and, yes, intelligent book. Hawkins has persuaded me that "intelligent" machines are very likely in our near future. And I'm sure they will be very helpful and will have no reason to do anything mean to us, their intellectual inferiors.
If they do turn on us, then, well, maybe God will help us.
December 17, 2020
This was complex reading on a complex topic from a long time ago, so the details are hazy.
Re-reading my highlights, I recall that some interesting hypotheses were how the brain is more like a very deep hierarchy of simple components repeated umpteenth times than some impossibly complex thing. A recursive design, such that the lower-level parts of the brain can perform the same computation, more or less, that gets transmitted to higher and higher levels of the brain which are actually identical-looking components performing the same calculations that nonetheless coalesce into higher and higher-level abstractions until it reaches our conscious mind.
It was very interesting to read.
Now, a year later, and having read about and practiced meditation, I would say this is not a bad hypothesis. The mind does form higher-level concepts from lower-level sensory data, and this composition goes pretty deep. Part of the point of meditation is developing the ability to look past the higher level concept, e.g. "the sound of a car revving up" or "the smell of coffee" to the raw sensory information itself.
Anyways, I'll probably have to re-read this to really solidify its ideas in my mind. Part of a deeper dive into intelligence and AI.
Re-reading my highlights, I recall that some interesting hypotheses were how the brain is more like a very deep hierarchy of simple components repeated umpteenth times than some impossibly complex thing. A recursive design, such that the lower-level parts of the brain can perform the same computation, more or less, that gets transmitted to higher and higher levels of the brain which are actually identical-looking components performing the same calculations that nonetheless coalesce into higher and higher-level abstractions until it reaches our conscious mind.
It was very interesting to read.
Now, a year later, and having read about and practiced meditation, I would say this is not a bad hypothesis. The mind does form higher-level concepts from lower-level sensory data, and this composition goes pretty deep. Part of the point of meditation is developing the ability to look past the higher level concept, e.g. "the sound of a car revving up" or "the smell of coffee" to the raw sensory information itself.
Anyways, I'll probably have to re-read this to really solidify its ideas in my mind. Part of a deeper dive into intelligence and AI.
November 2, 2017
Sadly, I found him to be arrogant. Normally I don't mind this, but it felt so indulgent it took away from the content.
Even the narrator had a tough time regurgitating his words. I'm pretty sure he was encoding "h e l p ... m e" in the duration of his spacing of words.
The good news is I treated it like a self help book as it drilled into my head 107 ways to stroke one's own ego.
Example "I'm not an expert in theoretical Atmospheric Astrophysics" yeah, no shit Sherlock.
He must have said "I'm not an expert" about 12 times in the book. But he only says it following pointing out they were wrong about something (that there is no proof they are wrong about), and making up some example where he is "right."
The whole book could be summed up as follows:
- I'm famous for creating Graffiti for the Palm Pilot, which was actually a hack because I could not write a good handwriting recognition formula like Apple had years before me on the Newton.
- I'm going to tell you 30 ways the brain is not like a computer, and a computer is not like a brain, while undermining my own points giving examples of other smarter researchers that keep chipping away at this very same problem and actually showing computers can in fact model exactly what the brain is doing a little better each year.
Even the narrator had a tough time regurgitating his words. I'm pretty sure he was encoding "h e l p ... m e" in the duration of his spacing of words.
The good news is I treated it like a self help book as it drilled into my head 107 ways to stroke one's own ego.
Example "I'm not an expert in theoretical Atmospheric Astrophysics" yeah, no shit Sherlock.
He must have said "I'm not an expert" about 12 times in the book. But he only says it following pointing out they were wrong about something (that there is no proof they are wrong about), and making up some example where he is "right."
The whole book could be summed up as follows:
- I'm famous for creating Graffiti for the Palm Pilot, which was actually a hack because I could not write a good handwriting recognition formula like Apple had years before me on the Newton.
- I'm going to tell you 30 ways the brain is not like a computer, and a computer is not like a brain, while undermining my own points giving examples of other smarter researchers that keep chipping away at this very same problem and actually showing computers can in fact model exactly what the brain is doing a little better each year.
December 20, 2021
What a book! I am blown away by the audacity and thoughtful ideas that he presents. As a cognitive neuroscience student, this was probably the most eye-opening book I’ve ever read about the brain.
I won’t spoil it, but I can say that the main idea is that Hawkins tries to explain how the cortex creates invariant representations and how this relates to intelligence.
Pros:
- Well-written with impactful interesting ideas
- Not too difficult to comprehend, but rich in details
- Ambitious and engaging
- Quick read
Cons
- Poor theories/ideas about consciousness. I don’t think he captured the idea of creativity either, but I might be wrong there.
- Too many predictions of how we can use the information on cortex operations to develop future machines.
All in all, a solid thought-provoking book. 4/5
I won’t spoil it, but I can say that the main idea is that Hawkins tries to explain how the cortex creates invariant representations and how this relates to intelligence.
Pros:
- Well-written with impactful interesting ideas
- Not too difficult to comprehend, but rich in details
- Ambitious and engaging
- Quick read
Cons
- Poor theories/ideas about consciousness. I don’t think he captured the idea of creativity either, but I might be wrong there.
- Too many predictions of how we can use the information on cortex operations to develop future machines.
All in all, a solid thought-provoking book. 4/5
Displaying 1 - 30 of 506 reviews