What do you think?
Rate this book
AIQ: How People and Machines are Smarter Together
“There comes a time in the life of a subject when someone steps up and writes the book about it. AIQ explores the fascinating history of the ideas that drive this technology of the future and demystifies the core concepts behind it; the result is a positive and entertaining look at the great potential unlocked by marrying human creativity with powerful machines.” —Steven D. Levitt, bestselling co-author of Freakonomics
From leading data scientists Nick Polson and James Scott, what everyone needs to know to understand how artificial intelligence is changing the world and how we can use this knowledge to make better decisions in our own lives.
Dozens of times per day, we all interact with intelligent machines that are constantly learning from the wealth of data now available to them. These machines, from smart phones to talking robots to self-driving cars, are remaking the world in the 21st century in the same way that the Industrial Revolution remade the world in the 19th century.
AIQ is based on a simple if you want to understand the modern world, then you have to know a little bit of the mathematical language spoken by intelligent machines. AIQ will teach you that language—but in an unconventional way, anchored in stories rather than equations.
You will meet a fascinating cast of historical characters who have a lot to teach you about data, probability, and better thinking. Along the way, you'll see how these same ideas are playing out in the modern age of big data and intelligent machines—and how these technologies will soon help you to overcome some of your built-in cognitive weaknesses, giving you a chance to lead a happier, healthier, more fulfilled life.
From leading data scientists Nick Polson and James Scott, what everyone needs to know to understand how artificial intelligence is changing the world and how we can use this knowledge to make better decisions in our own lives.
Dozens of times per day, we all interact with intelligent machines that are constantly learning from the wealth of data now available to them. These machines, from smart phones to talking robots to self-driving cars, are remaking the world in the 21st century in the same way that the Industrial Revolution remade the world in the 19th century.
AIQ is based on a simple if you want to understand the modern world, then you have to know a little bit of the mathematical language spoken by intelligent machines. AIQ will teach you that language—but in an unconventional way, anchored in stories rather than equations.
You will meet a fascinating cast of historical characters who have a lot to teach you about data, probability, and better thinking. Along the way, you'll see how these same ideas are playing out in the modern age of big data and intelligent machines—and how these technologies will soon help you to overcome some of your built-in cognitive weaknesses, giving you a chance to lead a happier, healthier, more fulfilled life.
288 pages, Kindle Edition
First published May 15, 2018
205 people are currently reading
1299 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 121 reviews
October 7, 2020
I bought this one during an Amazon sale on a whim. Generally with technology books, you must read them sooner rather than later, else you run the risk of them being outdated. Technology advances at breakneck speed.
This book mixes history with technology to instruct in how AI works and how and why humans are still required...at least so far. I loved the chapter showcasing Grace Hopper as the inventor of the subroutine for programming (among many other things). A mixture of AI learning and commentary on the vast accumulations of data how we use it and what/why we need it and AIs. Also a primer on statistics and an intro to "deep learning" (how AIs think). Very well done and a fascinating blend of history that lead to the foundations of the technology and the mathematical connections to the technology. Ever hear of Isaac Newton's "Trial of Pyx" during his role as the warden of the Royal Mint? This was the history used to instruct on the importance of understanding variance. How about an AI prediction for trends among contraception that said that in 10 yrs on the pill a person would have a 60% chance of getting pregnant. That chapter was an instruction on the dangers of a model based upon the wrong data. An example of why AI currently needs humans to interpret findings. An AI can spot trends mining mountains of data, but humans have to create the model, evaluate and refine it based upon findings.
4+ Stars
Read on kindle
This book mixes history with technology to instruct in how AI works and how and why humans are still required...at least so far. I loved the chapter showcasing Grace Hopper as the inventor of the subroutine for programming (among many other things). A mixture of AI learning and commentary on the vast accumulations of data how we use it and what/why we need it and AIs. Also a primer on statistics and an intro to "deep learning" (how AIs think). Very well done and a fascinating blend of history that lead to the foundations of the technology and the mathematical connections to the technology. Ever hear of Isaac Newton's "Trial of Pyx" during his role as the warden of the Royal Mint? This was the history used to instruct on the importance of understanding variance. How about an AI prediction for trends among contraception that said that in 10 yrs on the pill a person would have a 60% chance of getting pregnant. That chapter was an instruction on the dangers of a model based upon the wrong data. An example of why AI currently needs humans to interpret findings. An AI can spot trends mining mountains of data, but humans have to create the model, evaluate and refine it based upon findings.
"That’s because there’s no algorithm yet invented that can propose, test, and justify its own assumptions. Algorithms just do exactly what they’re told".The book also went into detail about the medical applications using Florence Nightingale's historic attention to detail to describe how information can be used to help diagnose and cure medical issues. Also gives breakdown of the logic that goes into creating a driverless car. If you drive for Lyft, this technology is coming fast. More voice/speech recognition technology and the logic behind it. The authors (both data scientists with a background in statistics) respected their readers enough not to dumb it down too much, but also acknowledged that not every one is a data scientist nor do we want to be (though maybe). Don't let the uninspiring cover and/or title fool you. This was another random purchase during the kindle sale that turned out to be a winner (though I am a technology geek). I really enjoyed this!!
4+ Stars
Read on kindle
July 21, 2018
This is by far the clearest and realistic book about Artificial Intelligence that I have ever read to date.
Rather than testing AI like a black box and readers like kindergarten kids (you won’t understand anyway), Polson explains in clear English how AI works. Basically AI uses Bayes’ theorem. Start with base rates, and adjust when new data comes so you get better and better. Even though a lot of the AI has proprietary algorithms (trade secrets), the principles are the same.
Each AI algorithm only does 1 thing well pretty much like an app. Polson thinks that worrying about AI getting sentience is like worrying about our space vehicles approaching a black hole when we have just invented the first aeroplane. He categorically states that no matter how smart AI looks, it does not really understand anything and is just crunching numbers.
So what makes good AI?
1. A large data set and lots of computing power.
2. Smart AI researchers who ask the right questions and tries to eliminate biases. If we just take raw data which already incorporate biases (e.g. women are paid less for the same job), then AI will just propagate the same bias.
3. Continual improvement because AI algorithm grows old pretty quickly. He gives the cautionary tale of Google’s (now stopped) Flu Epidemic detection app. It overestimated flu severity and has been scrapped. This is because by prompting searches with search suggestions, it changes the search itself.
I especially like the clear explanation on speech recognition by ‘word vector’. Reading this book makes you realise that Skynet is still very far away.
Rather than testing AI like a black box and readers like kindergarten kids (you won’t understand anyway), Polson explains in clear English how AI works. Basically AI uses Bayes’ theorem. Start with base rates, and adjust when new data comes so you get better and better. Even though a lot of the AI has proprietary algorithms (trade secrets), the principles are the same.
Each AI algorithm only does 1 thing well pretty much like an app. Polson thinks that worrying about AI getting sentience is like worrying about our space vehicles approaching a black hole when we have just invented the first aeroplane. He categorically states that no matter how smart AI looks, it does not really understand anything and is just crunching numbers.
So what makes good AI?
1. A large data set and lots of computing power.
2. Smart AI researchers who ask the right questions and tries to eliminate biases. If we just take raw data which already incorporate biases (e.g. women are paid less for the same job), then AI will just propagate the same bias.
3. Continual improvement because AI algorithm grows old pretty quickly. He gives the cautionary tale of Google’s (now stopped) Flu Epidemic detection app. It overestimated flu severity and has been scrapped. This is because by prompting searches with search suggestions, it changes the search itself.
I especially like the clear explanation on speech recognition by ‘word vector’. Reading this book makes you realise that Skynet is still very far away.
September 11, 2019
This book is not really about how artificial intelligence works, but more a collection of interesting historical tidbits and examples of current AI use in the real world. As such, it is quite entertaining, but not particularly illuminating.
AI in its current form performs statistical analysis on vast amounts of data to discover patterns. With the explosion of the internet and mobile devices, the data collected about anything has grown huge. This enabled the creation of self-training algorithms. They rely on some basic assumptions, then gather data, then adjust the previous assumptions with new predictions. As the dataset grows, predictions get better. This is how they managed to get speech recognition and translation much better - not by encoding rules, but discovering patterns in the growing set of recorded speech of many people across networks.
This is not the glamorous android assistant of sci-fi novels, but a bunch of number-crunching algorithms chained together, built on top of massive cloud computing power. But, it is way more pervasive in changing our lives. Polson scans the fields of usage from Netflix recommendations through self-driving cars to healthcare. He also talks of the pitfalls of wrong assumptions, letting models go out of date, fears from AI taking over (sorry, Elon Musk, you are a bit of a nutjob), cultural and moral implications.
If you want to learn about AI, this is not the book. It is however an entertaining collection of stories related to AI technology. It was not what I expected, but I liked it.
AI in its current form performs statistical analysis on vast amounts of data to discover patterns. With the explosion of the internet and mobile devices, the data collected about anything has grown huge. This enabled the creation of self-training algorithms. They rely on some basic assumptions, then gather data, then adjust the previous assumptions with new predictions. As the dataset grows, predictions get better. This is how they managed to get speech recognition and translation much better - not by encoding rules, but discovering patterns in the growing set of recorded speech of many people across networks.
This is not the glamorous android assistant of sci-fi novels, but a bunch of number-crunching algorithms chained together, built on top of massive cloud computing power. But, it is way more pervasive in changing our lives. Polson scans the fields of usage from Netflix recommendations through self-driving cars to healthcare. He also talks of the pitfalls of wrong assumptions, letting models go out of date, fears from AI taking over (sorry, Elon Musk, you are a bit of a nutjob), cultural and moral implications.
If you want to learn about AI, this is not the book. It is however an entertaining collection of stories related to AI technology. It was not what I expected, but I liked it.
June 3, 2018
AI pervades the fabric of our society and its influence upon our daily lives is expanding exponentially, literally "changing the world one smartphone at a time."
In AIQ, Nicholas Polson and James Scott who are bona fide data scientists, urge us to not only embrace this technological revolution, but also to give remembrance to mostly unheralded historical figures from a much earlier time, whose mathematical theories and data-analysis forms the very bedrock of AI technology and development in its current sophisticated state. From Facebook to self-navigation cars to seeking nuclear warheads, you will be amazed at how centuries old math is still informing the math behind modern technology.
The surge in AI development in the last decade is owed to three enabling technological forces that have supercharged AI into a new age. Namely, the exponential growth in the speed of computers called Moore's law, the explosive growth in the amount of data generally available and cloud computing, all resulting in the minimising of cost and radically upscaling the computing function of powerful algorithms.
What truly enhances the reading and intellectual pleasure of AIQ, is the fact that the authors didn't seek to dumb down the subject matter, but rather took their readers by the hand and guided them in a very accessible way through the myriad of math and theory, so that we can truly appreciate and understand this complex topic in a meaningful way, with illuminating real life examples to illustrate their technical points and ultimately help readers close out AIQ with a better understanding of the digital world around us.
In AIQ, Nicholas Polson and James Scott who are bona fide data scientists, urge us to not only embrace this technological revolution, but also to give remembrance to mostly unheralded historical figures from a much earlier time, whose mathematical theories and data-analysis forms the very bedrock of AI technology and development in its current sophisticated state. From Facebook to self-navigation cars to seeking nuclear warheads, you will be amazed at how centuries old math is still informing the math behind modern technology.
The surge in AI development in the last decade is owed to three enabling technological forces that have supercharged AI into a new age. Namely, the exponential growth in the speed of computers called Moore's law, the explosive growth in the amount of data generally available and cloud computing, all resulting in the minimising of cost and radically upscaling the computing function of powerful algorithms.
What truly enhances the reading and intellectual pleasure of AIQ, is the fact that the authors didn't seek to dumb down the subject matter, but rather took their readers by the hand and guided them in a very accessible way through the myriad of math and theory, so that we can truly appreciate and understand this complex topic in a meaningful way, with illuminating real life examples to illustrate their technical points and ultimately help readers close out AIQ with a better understanding of the digital world around us.
September 24, 2018
I love this book. The authors literally describe AI without using math! It’s incredible. There’s a nice bit of quantitative expounding, but each chapter mostly focuses on the actuality, history, and practical application of big AI concepts, such as Bayesian stats, deep learning, and neural networks. I really loved the chapters on Grace Hopper and Florence Nightingale :)
September 5, 2018
Context is the biggest motivator there is to learn something. This book is on point in that regard.First let us learn about how AI came to be, what it really does and how it does it. Then jump into writing code and working fancy math.
August 18, 2023
This a 3 star for non-nerds and possibly a 4 for nerds. Story wise I thought Polson did a good job showing a variety of ways data science and statistics currently impact our lives and some of the future possibilities. The full title is acurately descriptive, this is a book less about AI, although it has some discussion and a ton about data mining, high level statistics and crunching numbers. The true nerds among my vast number of followers will note that, that is all AI is, filtering through data looking for a meaningfull signal. Machine intelligence accomplishes this by extrapolating or substituting different rules than the ones initially supplied.
I enjoy the pieces discussion that fact that although these statistical methods have been around forever, it has only recently that the combination of massive amounts of data and the computing power to deal with it has become available to really make it useful.
The way bias is introduced was also well done.
I don't think this is a recommend to everyone for me, it is a bit dry and he gets lost in the math at times but if you are already interested in the subject I think it is a solid book.
I enjoy the pieces discussion that fact that although these statistical methods have been around forever, it has only recently that the combination of massive amounts of data and the computing power to deal with it has become available to really make it useful.
The way bias is introduced was also well done.
I don't think this is a recommend to everyone for me, it is a bit dry and he gets lost in the math at times but if you are already interested in the subject I think it is a solid book.
November 29, 2018
Pretty interesting, but really mostly enjoyable as a launch pad for other interesting stuff.
It's really hard to cover this subject matter in a mass-market book. For example, Three Blue One Brown has a far superior explanation for how neural networks function, through his incredible YouTube series
Theres some nice stories to explain statistical concepts like survivorship bias, Bayes' theorem, factor analysis, the base rate fallacy, and SLAM, but you're better off just with the wikipedia pages.
One particular story stood out for me, which involved Isaac Newton not understanding the implications of the Standard Error of the Mean (discovered by De Moivre), and subsequently losing the British government enormous amounts of money to coin clipping.
De Moivre's idea was that averages of samples should vary less than the entire population, because the highs and lows in a sample will average out. The variability between averages should equal the variability of the population divided by the square root of the sample size. Newton didn't understand this, and thought when he pulled samples of coins to average, he thought he could allow for the accepted variability for the entire population. I found this pdf which has other cool examples of this statistical fallacy.
It's really hard to cover this subject matter in a mass-market book. For example, Three Blue One Brown has a far superior explanation for how neural networks function, through his incredible YouTube series
Theres some nice stories to explain statistical concepts like survivorship bias, Bayes' theorem, factor analysis, the base rate fallacy, and SLAM, but you're better off just with the wikipedia pages.
One particular story stood out for me, which involved Isaac Newton not understanding the implications of the Standard Error of the Mean (discovered by De Moivre), and subsequently losing the British government enormous amounts of money to coin clipping.
De Moivre's idea was that averages of samples should vary less than the entire population, because the highs and lows in a sample will average out. The variability between averages should equal the variability of the population divided by the square root of the sample size. Newton didn't understand this, and thought when he pulled samples of coins to average, he thought he could allow for the accepted variability for the entire population. I found this pdf which has other cool examples of this statistical fallacy.
December 30, 2018
This is one of the most approachable math or computer science books I’ve ever read, and on a super important topic. Everyone who wants to be smarter should put this at the top of their list.
Starts as presenting itself as a very approachable general reader introduction to the broadly applicable value of machine learning. Entertaining writing, even some hilarious phrasing. e.g. “you can see parallax for yourself […] raise your index finger, or perhaps a different finger if you’re cursing us for all the math..” (p52). “ but this is a textbook case of economists naming obvious things after other economists.” (p152)
Chapter one starts very strong, explaining conditional probably vividly and super memorably with Abraham Wald and his WWII airplane survival analyses.
“Smart people who care about the world simply must know more about AI. That’s one reason we wrote this book.” (p39) And it is a great reason why you should read it!!
Each story herein was new to me, and they were, with just one trivial exception, all interesting and memorable.
Excellent conclusion (p238): “Now imagine a world where were actually smart about our efforts – a world where we put the right experts and the right legal protections in the right places, and where we’re eternally vigilant about the biases and assumptions of our algorithms. In that world our decision-making protocols could become radically better than the bias-riddled ones we have now. [...] our collective vision and technology have reached a point where we can successfully teach machines to drive a car, predict kidney disease, and carry on a conversation. We can certainly teach those machines to play fair. They might even teach us. [...] when it comes to the important decisions in life, we can and should combine artificial intelligence with human insight and human values. All it takes is people machines working together.”
Starts as presenting itself as a very approachable general reader introduction to the broadly applicable value of machine learning. Entertaining writing, even some hilarious phrasing. e.g. “you can see parallax for yourself […] raise your index finger, or perhaps a different finger if you’re cursing us for all the math..” (p52). “ but this is a textbook case of economists naming obvious things after other economists.” (p152)
Chapter one starts very strong, explaining conditional probably vividly and super memorably with Abraham Wald and his WWII airplane survival analyses.
“Smart people who care about the world simply must know more about AI. That’s one reason we wrote this book.” (p39) And it is a great reason why you should read it!!
Each story herein was new to me, and they were, with just one trivial exception, all interesting and memorable.
Excellent conclusion (p238): “Now imagine a world where were actually smart about our efforts – a world where we put the right experts and the right legal protections in the right places, and where we’re eternally vigilant about the biases and assumptions of our algorithms. In that world our decision-making protocols could become radically better than the bias-riddled ones we have now. [...] our collective vision and technology have reached a point where we can successfully teach machines to drive a car, predict kidney disease, and carry on a conversation. We can certainly teach those machines to play fair. They might even teach us. [...] when it comes to the important decisions in life, we can and should combine artificial intelligence with human insight and human values. All it takes is people machines working together.”
November 1, 2018
See my review here: https://www.bradrevell.com/2018/10/ai...
Three key takeaways from the book:
1. A domain specific illusion of intelligent behaviour is what happens when you chain a lot of algorithms together in a clever way. This is essentially what AI, Machine Learning, Deep Learning is. There is a lot of sales speak on these terms however “domain” and “specific” is absolutely key aspects to the AI definition; now and for the foreseeable future.
2. In AI, the role of the programmer isn’t to tell the algorithm what to do. It’s to tell the algorithm how to train itself what to do, using data and the rules of probability.
3. To a learning machine, “personalisation” means “conditional probability”.
Three key takeaways from the book:
1. A domain specific illusion of intelligent behaviour is what happens when you chain a lot of algorithms together in a clever way. This is essentially what AI, Machine Learning, Deep Learning is. There is a lot of sales speak on these terms however “domain” and “specific” is absolutely key aspects to the AI definition; now and for the foreseeable future.
2. In AI, the role of the programmer isn’t to tell the algorithm what to do. It’s to tell the algorithm how to train itself what to do, using data and the rules of probability.
3. To a learning machine, “personalisation” means “conditional probability”.
May 12, 2019
Interesting examples and powerful storytelling!
Particularly loved the chapter on Bayes' theorem (search operation for USS Scorpion) and the chapter on Sampling (trial of the Pyx) in which the authors mercilessly go after Isaac Newton. Overall, a great read (despite some dramatization here and there)
Particularly loved the chapter on Bayes' theorem (search operation for USS Scorpion) and the chapter on Sampling (trial of the Pyx) in which the authors mercilessly go after Isaac Newton. Overall, a great read (despite some dramatization here and there)
March 9, 2019
I give these authors great credit for actually trying to explain how it all works.
September 17, 2018
No que toca a livros que sirvam de introdução à Inteligência Artificial, The Master Algorythm de Pedro Domingos é a grande referência contemporânea. No entanto, apesar de profundo e abrangente, esse não é um livro que se compadeça dos leitores mais leigos. Denso, pesado, dá-nos de facto um panorama do estado da arte em IA e possibilidades futuras, mas não é fácil de levar até ao fim. Algo que não acontece neste AIQ.
Claro que AIQ não tem o objetivo de tentar mostrar como se podem unificar as diferentes vertentes da IA. Este pretende ser, e consegue-o, uma introdução aprofundada aos domínios da IA. É profundo mas não é denso, sendo compreensível pelos comuns mortais que não matutam profundamente em lógica e matemática probabilística no dia a dia. Um pormenor intrigante é que cada capítulo explora uma vertente da IA indo à história da ciência. É essa a grande lição deste livro, o percebermos que as técnicas e a matemática por detrás dos correntes avanços na Inteligência Artificial têm raízes na astronomia, matemática e outras ciências do iluminismo ao século XX.
É curioso pensar que o motor da IA, a matemática, algoritmia e probabilística, antecede e muito a corrente revolução em Inteligência Artificial. O que permitiu a sua explosão foi a combinação tripla de estruturas matemáticas, hardware acessível e potente, e a vastidão de dados possibilitados pela internet. Combinado algoritmos com processamento e big data, podemos tirar partido do poder da IA.
É de notar que os autores depressa desmontam o mito da IA como uma consciência super-inteligente. São mais terra a terra, mostram-nos que os avanços nestas tecnologia acontecem na forma restrita. Quer sejam os algoritmos de predição de produtos com que interagimos nas lojas online (ou no Netflix), algoritmos de reconhecimento de imagem, produtos como o Alexa, que agrupam tecnologias de processamento de linguagem natural com algoritmos de pesquisa e de predição com base no perfil do utilizador, sistemas de navegação por gps, processamento de dados financeiros (na banca, seguros, ou no IMTT, que está a usar uma ferramenta de IA para processar multas rodoviárias). Não são seres artificiais inteligentes, são ferramentas poderosas em aplicações muito específicas. São inteligentes, de formas muito específicas. Comportamentos inteligentes não significam consciência, na natureza encontramos exemplos disso, exames de abelhas ou formigas apresentam comportamentos complexos e estigmérgicos.
AIQ não nos mergulha tão a fundo na IA que poderemos temer o afogamento. Mantém-se leve, acessível, e dá-nos uma visão surpreendente e profunda da evolução e impactos de uma tecnologia que já está a influenciar a forma como vivemos.
Claro que AIQ não tem o objetivo de tentar mostrar como se podem unificar as diferentes vertentes da IA. Este pretende ser, e consegue-o, uma introdução aprofundada aos domínios da IA. É profundo mas não é denso, sendo compreensível pelos comuns mortais que não matutam profundamente em lógica e matemática probabilística no dia a dia. Um pormenor intrigante é que cada capítulo explora uma vertente da IA indo à história da ciência. É essa a grande lição deste livro, o percebermos que as técnicas e a matemática por detrás dos correntes avanços na Inteligência Artificial têm raízes na astronomia, matemática e outras ciências do iluminismo ao século XX.
É curioso pensar que o motor da IA, a matemática, algoritmia e probabilística, antecede e muito a corrente revolução em Inteligência Artificial. O que permitiu a sua explosão foi a combinação tripla de estruturas matemáticas, hardware acessível e potente, e a vastidão de dados possibilitados pela internet. Combinado algoritmos com processamento e big data, podemos tirar partido do poder da IA.
É de notar que os autores depressa desmontam o mito da IA como uma consciência super-inteligente. São mais terra a terra, mostram-nos que os avanços nestas tecnologia acontecem na forma restrita. Quer sejam os algoritmos de predição de produtos com que interagimos nas lojas online (ou no Netflix), algoritmos de reconhecimento de imagem, produtos como o Alexa, que agrupam tecnologias de processamento de linguagem natural com algoritmos de pesquisa e de predição com base no perfil do utilizador, sistemas de navegação por gps, processamento de dados financeiros (na banca, seguros, ou no IMTT, que está a usar uma ferramenta de IA para processar multas rodoviárias). Não são seres artificiais inteligentes, são ferramentas poderosas em aplicações muito específicas. São inteligentes, de formas muito específicas. Comportamentos inteligentes não significam consciência, na natureza encontramos exemplos disso, exames de abelhas ou formigas apresentam comportamentos complexos e estigmérgicos.
AIQ não nos mergulha tão a fundo na IA que poderemos temer o afogamento. Mantém-se leve, acessível, e dá-nos uma visão surpreendente e profunda da evolução e impactos de uma tecnologia que já está a influenciar a forma como vivemos.
August 10, 2019
It explains in details the methods of Artificial Intelligence and describes the statistics behind them. Also it presents real life applications and the future of this promising field.
August 18, 2024
This is hands down the best book about AI I've read so far. Highly recommend.
June 10, 2024
This was good but not great; worn-out staple examples, an ok writing style but a bit dull; the subject matter could've been fleshed out more and presented in a more upbeat, refreshing tone; now you're left with the feeling that with the emergence of AI very great and very terrible things can happen, and that's it.
November 4, 2018
I read this book after seeing a review about it in the Wall Street Journal. Then, after I started it, Wes Gray (Alpha Architect) recommended AIQ as the best book to get an understanding of artificial intelligence in a book. Having finished it, I agree. Obviously, one 250 page book can't cover everything, but they build a solid foundation. However, I don't have a background in computer science or coding, so I could be wrong.
To summarize what I took away from the book:
1) Conditional probability has been the key to personalizing and advancing AI over the past two decades. The computing power needed to determine such probabilities were not available until the 1990's and the massive datasets needed to analyze weren't available until the internet exploded.
2) Modeling, in AI, is based on prediction rules. The prediction rule for a linear equation is y = mx +b, while the prediction rule for a photo, can have up to 400,000 variables (which is why massive datasets are required to be able to derive such equations without having overfitting).
3) the prediction equations, using massive datasets and massive variables, are referred to as neural networks
4) For autonomous vehicles, a major challenge is simultaneous localization and mapping (SLAM). Apparently, humans ability to walk into a room and orient themselves is a skill that is not fully understood and is incredibly hard to model for vehicles.
5) Natural language processing has improved exponentially since 2010 because a rules-based approach to language has been abandoned. Instead, words are converted to numbers and assigned probabilities (word2vec). Assigning numerical values to words has resulted in applications like Alexa and Google Home.
There is a lot of interesting history in this book as well as a lot of great examples. They also cover specific examples of AI application (like in healthcare), but the most interesting part of the book (to me) was the explanation of prediction rules and challenges presented when implementing AI.
To summarize what I took away from the book:
1) Conditional probability has been the key to personalizing and advancing AI over the past two decades. The computing power needed to determine such probabilities were not available until the 1990's and the massive datasets needed to analyze weren't available until the internet exploded.
2) Modeling, in AI, is based on prediction rules. The prediction rule for a linear equation is y = mx +b, while the prediction rule for a photo, can have up to 400,000 variables (which is why massive datasets are required to be able to derive such equations without having overfitting).
3) the prediction equations, using massive datasets and massive variables, are referred to as neural networks
4) For autonomous vehicles, a major challenge is simultaneous localization and mapping (SLAM). Apparently, humans ability to walk into a room and orient themselves is a skill that is not fully understood and is incredibly hard to model for vehicles.
5) Natural language processing has improved exponentially since 2010 because a rules-based approach to language has been abandoned. Instead, words are converted to numbers and assigned probabilities (word2vec). Assigning numerical values to words has resulted in applications like Alexa and Google Home.
There is a lot of interesting history in this book as well as a lot of great examples. They also cover specific examples of AI application (like in healthcare), but the most interesting part of the book (to me) was the explanation of prediction rules and challenges presented when implementing AI.
July 4, 2018
Great introduction to the understanding of what artificial intelligence is and the enormous potential it has for making our lives better and smarter. Written for the non-scientist or mathematician this book explains what AI is, its evolution, where it is now and what it offers us going forward. For those with vague fears of computers taking over the world AIQ frames the reality of human control and benefits vs. pitfalls. Recommended.
July 18, 2018
This book is 90% fluff/backstory and 10% AI details.
Yeah sure some of the fluff was interesting, but it’s not AI.
And for that matter, the AI info was mostly Bayesian statistics.
Not a very broad approach. Not a very deep approach.
I know they were avoiding too much math, but they went overboard
Yeah sure some of the fluff was interesting, but it’s not AI.
And for that matter, the AI info was mostly Bayesian statistics.
Not a very broad approach. Not a very deep approach.
I know they were avoiding too much math, but they went overboard
December 13, 2018
Long winded explanation of a myopic view of AI, obsessed by analytics while ignoring pretty much everything else. Needs a good edit and new title. Not terrible though, for what it is.
March 9, 2020
una miniera di conoscenza. e una pietra miliare per capire il mondo moderno
March 7, 2020
I don't remember I have ever bought a book just because of the look of it. But this one, in Delhi airport few months ago, attracted me by the sheer fluorescent yellow cover and shining metal blue lettering; and I just bought it. The subject of AI is pretty close to my personal and professional interest and thought it wouldn't be a misfit in my library! Little did I know that it will e suchhh and intersting reading pleasure.
The authors narrative style, always picking up a story from the human history of different ages, and then linking the same with one modern marvel of artificial intelligence solution; is a delight to any reader.
For example, how he linked the great Florence Nightingale, the pioneer in modern scientific nursing, how she overcome so many social and personal obstacles to take up the cause of serving the ailing, and her fearless attitude of serving the war victims. But little did we know that she was also pioneer in enumerating the basic health parameters regularly for a patient; that help monitor the condition of the patient over time and how it helped doctors to make more wise, data driven decision. Then he links this story to the modern medicine and clinical methodologies; and extends this further to the health apps of today and tomorrow - which will monitor the vitals of a health or ailing person and predict the need for medical attention or lifestyle balancing.
Each chapter is full with such wonders. One must read through all to get the big picture of AI - how it is impacting all the aspects of our lives.
One small disappointment was that not all the examples he has mentioned are great AI examples or use cases, some are pretty simplistic and old by now. And there are much more advancement already taken place in many use cases which the author have missed. but this could be the case that we the industry insiders are privy to such cases and these are not common knowledge yet.
The authors narrative style, always picking up a story from the human history of different ages, and then linking the same with one modern marvel of artificial intelligence solution; is a delight to any reader.
For example, how he linked the great Florence Nightingale, the pioneer in modern scientific nursing, how she overcome so many social and personal obstacles to take up the cause of serving the ailing, and her fearless attitude of serving the war victims. But little did we know that she was also pioneer in enumerating the basic health parameters regularly for a patient; that help monitor the condition of the patient over time and how it helped doctors to make more wise, data driven decision. Then he links this story to the modern medicine and clinical methodologies; and extends this further to the health apps of today and tomorrow - which will monitor the vitals of a health or ailing person and predict the need for medical attention or lifestyle balancing.
Each chapter is full with such wonders. One must read through all to get the big picture of AI - how it is impacting all the aspects of our lives.
One small disappointment was that not all the examples he has mentioned are great AI examples or use cases, some are pretty simplistic and old by now. And there are much more advancement already taken place in many use cases which the author have missed. but this could be the case that we the industry insiders are privy to such cases and these are not common knowledge yet.
December 17, 2022
I actually finished this book quite sometime ago but I keep referring back to it. Anyone one with an interest in artificial intelligence and the impact it has on everyday life I feel will find it quite interesting.
March 8, 2024
This book did a great job at explaining AI in layman's terms. It laid a great foundation for the technical and ethical implications of artificial intelligence. While I usually find academic books boring, Polson and Scott kept me engaged and delivered the information concisely and with interesting anecdotes.
November 6, 2019
I thoroughly enjoyed reading this book. Nick & James have very successfully explained AI in a simple fashion and in the process have busted many myths and lies surrounding the concept. Even better, they have done so with humour. In doing so they have brought the subject down to the levels where non-technical people will be able to relate to it. I will highly recommend this book to everyone.
January 15, 2024
Good interesting book about ideas behind AI. Has little snippets of history and statistics that have influenced the field today, in a very simple to understand way. Would probs actually give 3.5 stars, but I’ll round up :).
August 26, 2019
I recommend this book to anyone who is even remotely interested in understanding how artificial intelligence is starting to transform business and society. The story-based writing style makes this rather intimidating topic much easier to comprehend too.
June 23, 2019
AIQ Review
A short and educational read explaining how Artificial Intelligence is nothing more than statistics with a ton of data on steroids. The book does a great job at raising the curtains that make AI seem so magical and mysterious. It also grounds all the misconceptions and hype around a Skynet type general AI. Computers are much better than humans at remembering things, processing large amounts of data, and doing repetitive tasks very quickly. Similarly, a hammer is much better than humans at hitting things without being hurt or damaged. Both of these are tools that benefit humans, and the fear of being taken over by the tools we built is overhyped science fiction.
As a software engineer, who works closely with machine learning scientists, and has gone through Andrew Ng’s Coursera ML course, I still got a ton of value out of this book. I would recommend it to anyone who wants to freshen up on their stats knowledge, or expand their knowledge of history. In particular, the book does a great job at correlating modern advancements and uses of ML/AI to scientific developments of the past 100 years.
One of the things I hadn’t realized is that the big data, machines learning and cloud computing boom started around 2010. I was already in college and literally saw this revolution happen before my eyes!
Over the past few decades, engineers/scientists in this field really made a lot of mathematical and scientific breakthroughs. However, after reading this book, my previous assumption were validated. Almost nobody does “real machine learning” nowadays. Since various frameworks (pytorch, tensorflow, keras, caffe, etc…) have commoditized the ability to build and train neural networks, feature extraction has become the most difficult part of machine learning. The process of data collection, data cleansing, data analysis, data pipeline, and deciding how to represent that data is all machine learning has really come to. Everything else is just trial and error along with throwing a ton of data (i.e. steroids) at the problem. The book does a great job at exemplifying this by referring to two letters: Big N (big number of independent data points) and Big D (a lot of detail in every data point).
"It may seem like we rely on depend smart machines for everything these days, but in reality, they depend on us a lot more."
==== Some cool facts from the book that stood out ====
Moore’s Law
One of the AI/ML advancement enablers in the past decade is Moore’s Law. However, it’s not just Moore’s Law in the context of processors, but also in the context of data quantity and the availability cloud compute. All of this, along with many other enables, all played a key role in AI’s hyper growth.
UBI
A common concern amongst many who are afraid of AI is the number of jobs (i.e. truck driving) it’ll make disappear. As a big Andrew Yang fan myself, and a big proponent of Universal Basic Income, I really connected with the authors’ similar viewpoin. I do not want technological advancement to be hindered by a fear of job loss, but am aware that it could lead to an increase in unemployment. For example, truck driving, which is the most popular job in the will most certainly be impact by the ubiquity of autonomous vehicles. I believe that as long as we put UBI policies in place, and provide the infrastructure for individuals to be retrained in a different career path, if they choose to, we can move forward without looking back.
Netflix
In 2007, Netflix offered $1MM to anyone who could beat their recommendation algorithm by more than 10%. A lot of teams came really close, but it took 2 years for a team to beat Netflix’s recommendation algorithm by 0.06%. Ironically, another team also managed to submit a winning solution exactly 19 minutes after the first.
House of Cards was the first Netflix original, and as we all know, a great success. The reason this is the case is because Netflix had gathered a lot of data on what kind of shows that appeal to the general public, and therefore knew it was going to be a success.
The authors discuss how traditional television networks waste hundreds of millions to pilot many different shows to see what sticks. Netflix’s data allows them to create successful shows with a very low probability of failure.
World War 2
Abraham Wald played a key role in mathematic and statistical developments during WWII.
The planes that were returning from battle had a lot of holes in the fuselage, so the generals thought of adding more armor to that section of the plane. However, Abraham correctly identified that we should make note of the planes that do not return, which were likely shot in the engine, and that is where more armor should be added.
The key point of comparison between the bombers returning and Netflix’s recommendation system is that both of these systems use conditional probability with large data sets and latent features. Abraham Wald didn’t know where the non-returning planes were getting hit, and Netflix doesn’t know viewer’s opinions on most movies because most users do not watch most movies.
After WWII, the US spent $17B in the early 60s to install microphones across the whole Atlantic. How cool is that!
Asia - Toilet Paper Theft
Individuals in Japan and China used to steal toilet paper from public spaces. Authorities tried to limit the toilet paper “per visit” to 6 sheets. While this slowed things down, the thieves were just walking around and returning to the same booths periodically. The final solution was to use facial recognition and only dispense toilet paper per person every X hours.
Physicists
Hubble used the pulsating star theory, and collected a lot of data, to prove that Andromeda was located in a different Galaxy, thereby disproving the “single Galaxy theory”.
Elina Berglund, who helped discover the Higgs boson, decided to use big data to kill the birth control pill. A smartphone app, along with input from the user concerning body temperature and menstrual cycle became highly accurate at determining when a woman is ovulating.
Moravec's paradox
From wikipedia: “It is comparatively easy to make computers exhibit adult level performance on intelligence tests or playing checkers, and difficult or impossible to give them the skills of a one-year-old when it comes to perception and mobility”
Bayes' theorem
At the basis of all probability analyses is Baye’s theorem, which is actually quite straight forward: the probability that an event A occurs given that another event B has already occurred is equal to the probability that the event B occurs given that A has already occurred multiplied by the probability of occurrence of event A and divided by the probability of occurrence of event B.
SLAM
At the time of writing the book, GPS is only accurate up to 5m. This is good for most uses cases but not for autonomous vehicles, drones, etc… This is where SLAM (simultaneous localization and mapping) prevails. It does introspection by collecting data from various censors (color, grayscale, IMUs, depth, etc..) and exterminates (i.e. predicts) the device’s location/pose.
Grace Hopper
I had always heard of the Grace Hopper conference for women, but never realized how influential she was in the world of computer science.
Hopper grew up in a military family and managed to become one of the first female naval officers.
She wars the first American computer expert after she was assigned to work on the Mark I at Harvard. After that, she started working on the UNIVAC, which was the first large scale electronic computer adopted by various corporations such as ADP or Dupont for database operations.
She was a pioneer in NLP, computer compilers, and played a key role in making sure that the American government adapt to using modern computer systems.
Coin Clipping
The book went into great depth about coin clipping in the 16th century. Due to the variability in coin weight, it was easy to clip “excess” coin on heavy coins and make a little bit of extra money. It took a while for Isaac Newton to develop the method of least squares, which was followed by the great recoinage, to avoid this sort of theft in the future.
Cool Applications of AI
- There is a smart knife that tells the surgeons whether they’re cutting/burning healthy or cancerous tissues. This is done by sending the burning smoke to a mass spectrometer, analyzing the data and feeding it into a prediction rule This takes 3 seconds and has a 100% accuracy rate.
- Ali baba ships products to nearby warhorses before items are purchased to guarantee faster delivery, since they can predict which products are going to be bought when and where!
Florence Nightingale
Florence nightingale was a nurse and data scientist. In the 1800s, she was driven by both maths and helping people. She used her skills and interests to built an efficient data driven hospitals/camps to help as many injured soldiers as possible. She was the first woman elected to the UK data analytics organization, and was truly a remarkable and influential individual.
Compounding
If something has a 0.9 chance of happening, the compounding rule states that the chance of it happen 10 times is 0.9^10. However, the compunding rule does not apply to population averages where there are a lot of lurking variables that need to be taken into consideration, and often leads to conclusion bias.
For example, if 90% of the population has a tails only coin, then 90% of all flips will always result in tails. However, the compounding rule will apply ONLY IF all the coins are fair.
A short and educational read explaining how Artificial Intelligence is nothing more than statistics with a ton of data on steroids. The book does a great job at raising the curtains that make AI seem so magical and mysterious. It also grounds all the misconceptions and hype around a Skynet type general AI. Computers are much better than humans at remembering things, processing large amounts of data, and doing repetitive tasks very quickly. Similarly, a hammer is much better than humans at hitting things without being hurt or damaged. Both of these are tools that benefit humans, and the fear of being taken over by the tools we built is overhyped science fiction.
As a software engineer, who works closely with machine learning scientists, and has gone through Andrew Ng’s Coursera ML course, I still got a ton of value out of this book. I would recommend it to anyone who wants to freshen up on their stats knowledge, or expand their knowledge of history. In particular, the book does a great job at correlating modern advancements and uses of ML/AI to scientific developments of the past 100 years.
One of the things I hadn’t realized is that the big data, machines learning and cloud computing boom started around 2010. I was already in college and literally saw this revolution happen before my eyes!
Over the past few decades, engineers/scientists in this field really made a lot of mathematical and scientific breakthroughs. However, after reading this book, my previous assumption were validated. Almost nobody does “real machine learning” nowadays. Since various frameworks (pytorch, tensorflow, keras, caffe, etc…) have commoditized the ability to build and train neural networks, feature extraction has become the most difficult part of machine learning. The process of data collection, data cleansing, data analysis, data pipeline, and deciding how to represent that data is all machine learning has really come to. Everything else is just trial and error along with throwing a ton of data (i.e. steroids) at the problem. The book does a great job at exemplifying this by referring to two letters: Big N (big number of independent data points) and Big D (a lot of detail in every data point).
"It may seem like we rely on depend smart machines for everything these days, but in reality, they depend on us a lot more."
==== Some cool facts from the book that stood out ====
Moore’s Law
One of the AI/ML advancement enablers in the past decade is Moore’s Law. However, it’s not just Moore’s Law in the context of processors, but also in the context of data quantity and the availability cloud compute. All of this, along with many other enables, all played a key role in AI’s hyper growth.
UBI
A common concern amongst many who are afraid of AI is the number of jobs (i.e. truck driving) it’ll make disappear. As a big Andrew Yang fan myself, and a big proponent of Universal Basic Income, I really connected with the authors’ similar viewpoin. I do not want technological advancement to be hindered by a fear of job loss, but am aware that it could lead to an increase in unemployment. For example, truck driving, which is the most popular job in the will most certainly be impact by the ubiquity of autonomous vehicles. I believe that as long as we put UBI policies in place, and provide the infrastructure for individuals to be retrained in a different career path, if they choose to, we can move forward without looking back.
Netflix
In 2007, Netflix offered $1MM to anyone who could beat their recommendation algorithm by more than 10%. A lot of teams came really close, but it took 2 years for a team to beat Netflix’s recommendation algorithm by 0.06%. Ironically, another team also managed to submit a winning solution exactly 19 minutes after the first.
House of Cards was the first Netflix original, and as we all know, a great success. The reason this is the case is because Netflix had gathered a lot of data on what kind of shows that appeal to the general public, and therefore knew it was going to be a success.
The authors discuss how traditional television networks waste hundreds of millions to pilot many different shows to see what sticks. Netflix’s data allows them to create successful shows with a very low probability of failure.
World War 2
Abraham Wald played a key role in mathematic and statistical developments during WWII.
The planes that were returning from battle had a lot of holes in the fuselage, so the generals thought of adding more armor to that section of the plane. However, Abraham correctly identified that we should make note of the planes that do not return, which were likely shot in the engine, and that is where more armor should be added.
The key point of comparison between the bombers returning and Netflix’s recommendation system is that both of these systems use conditional probability with large data sets and latent features. Abraham Wald didn’t know where the non-returning planes were getting hit, and Netflix doesn’t know viewer’s opinions on most movies because most users do not watch most movies.
After WWII, the US spent $17B in the early 60s to install microphones across the whole Atlantic. How cool is that!
Asia - Toilet Paper Theft
Individuals in Japan and China used to steal toilet paper from public spaces. Authorities tried to limit the toilet paper “per visit” to 6 sheets. While this slowed things down, the thieves were just walking around and returning to the same booths periodically. The final solution was to use facial recognition and only dispense toilet paper per person every X hours.
Physicists
Hubble used the pulsating star theory, and collected a lot of data, to prove that Andromeda was located in a different Galaxy, thereby disproving the “single Galaxy theory”.
Elina Berglund, who helped discover the Higgs boson, decided to use big data to kill the birth control pill. A smartphone app, along with input from the user concerning body temperature and menstrual cycle became highly accurate at determining when a woman is ovulating.
Moravec's paradox
From wikipedia: “It is comparatively easy to make computers exhibit adult level performance on intelligence tests or playing checkers, and difficult or impossible to give them the skills of a one-year-old when it comes to perception and mobility”
Bayes' theorem
At the basis of all probability analyses is Baye’s theorem, which is actually quite straight forward: the probability that an event A occurs given that another event B has already occurred is equal to the probability that the event B occurs given that A has already occurred multiplied by the probability of occurrence of event A and divided by the probability of occurrence of event B.
SLAM
At the time of writing the book, GPS is only accurate up to 5m. This is good for most uses cases but not for autonomous vehicles, drones, etc… This is where SLAM (simultaneous localization and mapping) prevails. It does introspection by collecting data from various censors (color, grayscale, IMUs, depth, etc..) and exterminates (i.e. predicts) the device’s location/pose.
Grace Hopper
I had always heard of the Grace Hopper conference for women, but never realized how influential she was in the world of computer science.
Hopper grew up in a military family and managed to become one of the first female naval officers.
She wars the first American computer expert after she was assigned to work on the Mark I at Harvard. After that, she started working on the UNIVAC, which was the first large scale electronic computer adopted by various corporations such as ADP or Dupont for database operations.
She was a pioneer in NLP, computer compilers, and played a key role in making sure that the American government adapt to using modern computer systems.
Coin Clipping
The book went into great depth about coin clipping in the 16th century. Due to the variability in coin weight, it was easy to clip “excess” coin on heavy coins and make a little bit of extra money. It took a while for Isaac Newton to develop the method of least squares, which was followed by the great recoinage, to avoid this sort of theft in the future.
Cool Applications of AI
- There is a smart knife that tells the surgeons whether they’re cutting/burning healthy or cancerous tissues. This is done by sending the burning smoke to a mass spectrometer, analyzing the data and feeding it into a prediction rule This takes 3 seconds and has a 100% accuracy rate.
- Ali baba ships products to nearby warhorses before items are purchased to guarantee faster delivery, since they can predict which products are going to be bought when and where!
Florence Nightingale
Florence nightingale was a nurse and data scientist. In the 1800s, she was driven by both maths and helping people. She used her skills and interests to built an efficient data driven hospitals/camps to help as many injured soldiers as possible. She was the first woman elected to the UK data analytics organization, and was truly a remarkable and influential individual.
Compounding
If something has a 0.9 chance of happening, the compounding rule states that the chance of it happen 10 times is 0.9^10. However, the compunding rule does not apply to population averages where there are a lot of lurking variables that need to be taken into consideration, and often leads to conclusion bias.
For example, if 90% of the population has a tails only coin, then 90% of all flips will always result in tails. However, the compounding rule will apply ONLY IF all the coins are fair.
August 26, 2020
Most of the books about artificial intelligence highlight such things as self-driving cars and facial recognition—the brilliant innovations in hardware and software by the engineers and coders who build the stuff and make it work. They dwell variously on the field’s potential and its impact on our lives today. This book, by two academic data scientists, instead drills down into the underlying logic of AI, the fuel that powers it, and the statistical rules that govern its operations. Because artificial intelligence is all about statistical probabilities. And AIQ by Nick Polson and James Scott truly fulfills the promise of its subtitle: How Artificial Intelligence Works and How We Can Harness Its Power for a Better World.
The statistical rules behind how artificial intelligence works
Surprised that statistics should be central to artificial intelligence? Polson and Scott explain. “Open up a blank email on your phone and try dictating a test phrase: ‘The weather report calls for rain, whether or not the reigning queen has an umbrella.’ If you’re a native English speaker and your phone runs iOS or Android, it will almost surely get the sentence right, without confusing weather/whether or rain/reign . . . The software knows that ‘whether’ and ‘reign’ are statistically more likely in some contexts, while ‘weather’ and ‘rain’ are more likely in others. This isn’t because your phone somehow understands the meaning of words. There’s no meaning involved, just a rich set of context-specific probabilities for basically every English word and phrase ever uttered on the internet.” And how does it know all that? Because the superfast computers on which the software runs have surveyed “basically every English word and phrase ever uttered on the internet.” (Yes, they can do that.)
The people who laid the foundation for artificial intelligence
One of the reasons this book reads so well is that the authors infuse each chapter with the story of an individual whose work—sometimes centuries ago—laid the foundation for what we now call artificial intelligence. And what people they are! The group includes some who are truly famous (Sir Isaac Newton and nurse-administrator Florence Nightingale) or near-famous (computer scientist Grace Hopper and baseball great Joe Dimaggio) as well as those you’ve probably never heard of (a Hungarian-American statistician named Abraham Wald and astronomer Henrietta Leavitt).
The surprising work of Florence Nightingale
Every one of these stories is fascinating, but the most in-depth of them features Florence Nightingale. “It turns out that when she wasn’t caring for soldiers, Nightingale was also a skilled data scientist who successfully convinced hospitals that they could improve health care using statistics. In fact, no other data scientist in history can claim to have saved so many lives as Florence Nightingale. In 1859, in honor of these achievements, she became the first woman ever elected to the U.K.’s Royal Statistical Society.” Surprising, isn’t it? (If not, go to the head of the class.)
It’s happening now, and it’s only the beginning
For the most part, in their effort to explain how artificial intelligence works, Polson and Scott dwell on its history and underlying logic. But they venture into the future as well. “Today?” they write. “Self-driving cars are just the start. Don’t forget about autonomous flying taxis, like the ones the government of Dubai has been testing since September of 2017. Or the autonomous iron mine run by Rio Tinto, in the middle of the Australian outback. Or the autonomous shipping terminal at the Port of Qingdao, in China—six enormous berths spanning two kilometers of coastline, 5.2 million shipping containers a year, hundreds of robot trucks and cranes, and nobody at the wheel.” So, is this a good thing, or bad? How many jobs will be lost through such projects? Polson and Scott don’t spend time on the question. You’ll need to read another book for perspective on the matter. This book will just help you understand how very logical it all is.
The statistical rules behind how artificial intelligence works
Surprised that statistics should be central to artificial intelligence? Polson and Scott explain. “Open up a blank email on your phone and try dictating a test phrase: ‘The weather report calls for rain, whether or not the reigning queen has an umbrella.’ If you’re a native English speaker and your phone runs iOS or Android, it will almost surely get the sentence right, without confusing weather/whether or rain/reign . . . The software knows that ‘whether’ and ‘reign’ are statistically more likely in some contexts, while ‘weather’ and ‘rain’ are more likely in others. This isn’t because your phone somehow understands the meaning of words. There’s no meaning involved, just a rich set of context-specific probabilities for basically every English word and phrase ever uttered on the internet.” And how does it know all that? Because the superfast computers on which the software runs have surveyed “basically every English word and phrase ever uttered on the internet.” (Yes, they can do that.)
The people who laid the foundation for artificial intelligence
One of the reasons this book reads so well is that the authors infuse each chapter with the story of an individual whose work—sometimes centuries ago—laid the foundation for what we now call artificial intelligence. And what people they are! The group includes some who are truly famous (Sir Isaac Newton and nurse-administrator Florence Nightingale) or near-famous (computer scientist Grace Hopper and baseball great Joe Dimaggio) as well as those you’ve probably never heard of (a Hungarian-American statistician named Abraham Wald and astronomer Henrietta Leavitt).
The surprising work of Florence Nightingale
Every one of these stories is fascinating, but the most in-depth of them features Florence Nightingale. “It turns out that when she wasn’t caring for soldiers, Nightingale was also a skilled data scientist who successfully convinced hospitals that they could improve health care using statistics. In fact, no other data scientist in history can claim to have saved so many lives as Florence Nightingale. In 1859, in honor of these achievements, she became the first woman ever elected to the U.K.’s Royal Statistical Society.” Surprising, isn’t it? (If not, go to the head of the class.)
It’s happening now, and it’s only the beginning
For the most part, in their effort to explain how artificial intelligence works, Polson and Scott dwell on its history and underlying logic. But they venture into the future as well. “Today?” they write. “Self-driving cars are just the start. Don’t forget about autonomous flying taxis, like the ones the government of Dubai has been testing since September of 2017. Or the autonomous iron mine run by Rio Tinto, in the middle of the Australian outback. Or the autonomous shipping terminal at the Port of Qingdao, in China—six enormous berths spanning two kilometers of coastline, 5.2 million shipping containers a year, hundreds of robot trucks and cranes, and nobody at the wheel.” So, is this a good thing, or bad? How many jobs will be lost through such projects? Polson and Scott don’t spend time on the question. You’ll need to read another book for perspective on the matter. This book will just help you understand how very logical it all is.
November 9, 2019
They have changed the title from the previous "How Artificial Intelligence Works and How We Can Harness Its Power for Good." The new title is better because the old subtitle claims too much--this book won't tell you how AI works. But it does give you a sense of how huge data sets and blazing fast computers are starting to come close to doing things that seemed beyond the reach of computers in the not at all distant past. I found the discussion of speech recognition to be particularly interesting. In the past scientists had tried to teach computers to "understand" or at least transcribe the language they hear based on a huge dictionary of rules for analyzing human speech. What we might call a grammar. Even with advances in computing power, this approach never really worked. Recently, though, a breakthrough has led to great advances. By the use of Bayesian probabilities and the analysis of huge oceans of language data, computers have been taught to recognize speech based on the probabilities of particular words occurring together in any given utterance. Once the words are converted to a set of probabilities based on collocations found in the emerging language data available now thanks to the WWW, the computers have become increasingly good at taking dictation. This is just one of the AI anecdotes told in the book, giving an interesting glimpse into the world of AI. If you have an interest in AI and you are not looking for a technical discussion of how it works, this book won't disappoint.
Displaying 1 - 30 of 121 reviews