What do you think?
Rate this book
数学之美
（第3版）
八年前，“数学之美”系列文章原刊载于谷歌黑板报，获得上百万次点击，得到读者高度评价。读者说，读了“数学之美”，才发现大学时学的数学知识，比如马尔可夫链、矩阵计算，甚至余弦函数原来都如此亲切，并且栩栩如生，才发现自然语言和信息处理这么有趣。
在纸本书的创作中，作者几乎把所 有文章都重写了一遍，为的是把高深的数学原理讲得更加通俗易懂，让非专业读者也能领略数学的魅力。读者通过具体的例子学到的是思考问题的方式 —— 如何化繁为简，如何用数学去解决工程问题，如何跳出固有思维不断去思考创新。
本书第一版荣获国家图书馆第八届文津图书奖。第二版增加了针对大数据和机器学习的内容。第三版增加了三章新内容，分别介绍当今非常热门的三个主题：区块链的数学基础，量子通信的原理，以及人工智能的数学极限。
八年前，“数学之美”系列文章原刊载于谷歌黑板报，获得上百万次点击，得到读者高度评价。读者说，读了“数学之美”，才发现大学时学的数学知识，比如马尔可夫链、矩阵计算，甚至余弦函数原来都如此亲切，并且栩栩如生，才发现自然语言和信息处理这么有趣。
在纸本书的创作中，作者几乎把所 有文章都重写了一遍，为的是把高深的数学原理讲得更加通俗易懂，让非专业读者也能领略数学的魅力。读者通过具体的例子学到的是思考问题的方式 —— 如何化繁为简，如何用数学去解决工程问题，如何跳出固有思维不断去思考创新。
本书第一版荣获国家图书馆第八届文津图书奖。第二版增加了针对大数据和机器学习的内容。第三版增加了三章新内容，分别介绍当今非常热门的三个主题：区块链的数学基础，量子通信的原理，以及人工智能的数学极限。
364 pages, Paperback
Published May 1, 2020
29 people are currently reading
171 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 - 12 of 12 reviews
March 24, 2022
1. Who should read this?
This book is important for computer scientist, mathematicians, statisticians, software engineers.
Because, it gives an outline of mathematical tools one needs to grasp solve problems in technology industry.
2. What is inside?
One cannot know everything but one needs to have an outline of what mathematical tools might be needed in the future to solve a problem. The broad scope of the book is to give the reader, an understanding of mathematics in Modern Computing especially through Author’s experience as a Research Scientist in Google.
The Purpose of this book is not to go into details of hidden algorithms behind a product.
We could reformulate the above as, the purpose is to give a gentle introduction to mathematical theories intuitively behind products, rather than software documentation or algorithms.
I took this book to help me categorize mathematics and how one uses them in industry. The Book is well-written and stories intertwined behind products inspire you.
The Chapter on Andrew Viterbi was the best for me.
Outline:
1- Words, Languages, Numbers & Information
2-NLP: From Rules to Statistics
3- Statistical Language Model
4-Word Segmentation
5-Hidden Markov Model
6-Quantifying Information
7-Jelinek and Modern Language processing
8-Boolean Algebra and Search Engines
9-Graph theory and Web Crawlers
10-Page Rank: Google’s ranking technology
11- Relevance in Web Search
12-Finite state machines and Dynamic Programming,
Google Maps & Navigation
13-Google’s Designer Ak-47, Dr. Amit Signal
14-Cosine and News Classification
15-Solving classification problem in text processing with matrices
16-Information fingerprint and application
17-Mathematical principles of Cryptography
18-Search Engine’s problem: Anti-Spam, authoritativeness
19-Importance of Mathematical models
20-Don’t put all your eggs in one basked: Principle of Maximum Entropy
21-Mathematical Principles of Chinese input method editors
22-Bloom Filters
23-Bayesian Network: Extension of Markov Chain
24-Conditional random fields, syntactic parsing
25-Andrew Viterbi and Viterbi algorithm
26-God’s algorithm: Expectation max Algorithm
27-Logistic Regression and Web Search Advertisement
28-Google Brain and Artificial neural network
29: Power of Big Data
Deus Vult,
Gottfried
This book is important for computer scientist, mathematicians, statisticians, software engineers.
Because, it gives an outline of mathematical tools one needs to grasp solve problems in technology industry.
2. What is inside?
One cannot know everything but one needs to have an outline of what mathematical tools might be needed in the future to solve a problem. The broad scope of the book is to give the reader, an understanding of mathematics in Modern Computing especially through Author’s experience as a Research Scientist in Google.
The Purpose of this book is not to go into details of hidden algorithms behind a product.
We could reformulate the above as, the purpose is to give a gentle introduction to mathematical theories intuitively behind products, rather than software documentation or algorithms.
I took this book to help me categorize mathematics and how one uses them in industry. The Book is well-written and stories intertwined behind products inspire you.
The Chapter on Andrew Viterbi was the best for me.
Outline:
1- Words, Languages, Numbers & Information
2-NLP: From Rules to Statistics
3- Statistical Language Model
4-Word Segmentation
5-Hidden Markov Model
6-Quantifying Information
7-Jelinek and Modern Language processing
8-Boolean Algebra and Search Engines
9-Graph theory and Web Crawlers
10-Page Rank: Google’s ranking technology
11- Relevance in Web Search
12-Finite state machines and Dynamic Programming,
Google Maps & Navigation
13-Google’s Designer Ak-47, Dr. Amit Signal
14-Cosine and News Classification
15-Solving classification problem in text processing with matrices
16-Information fingerprint and application
17-Mathematical principles of Cryptography
18-Search Engine’s problem: Anti-Spam, authoritativeness
19-Importance of Mathematical models
20-Don’t put all your eggs in one basked: Principle of Maximum Entropy
21-Mathematical Principles of Chinese input method editors
22-Bloom Filters
23-Bayesian Network: Extension of Markov Chain
24-Conditional random fields, syntactic parsing
25-Andrew Viterbi and Viterbi algorithm
26-God’s algorithm: Expectation max Algorithm
27-Logistic Regression and Web Search Advertisement
28-Google Brain and Artificial neural network
29: Power of Big Data
Deus Vult,
Gottfried
June 3, 2020
I feel refreshed.
The Beauty of Mathematics in Computer Science (BMCS) paints hope on a canvas which I have feared is but thinly veiling a research landscape saturated with papers consisting of throwing together integrations of pieces of exiting models or perhaps not much more than a simple increase in model size or parameter count without sufficiently supported insights. On such a bleak landscape, BMCS brings to life stories of researchers in natural language processing (NLP) and related fields with succinct but insightful explanations of the underlying mathematics. To the layman, "machine learning" (ML) is a foreign term, the all-consuming monster lurking in the corner where only those who have aced their high school and college math classes may approach. BMCS demonstrates that "machine learning" is no more than 10 sheets of book-sized paper filled with pictures, rightfully standing up against those who would rather use nebulous terms to show off their intelligence rather than attempting an honest explanation to a peer.
By no means is BMCS is meant to substitute a deeper dive into any of its topics. In Wu's struggle against his various NDAs, he makes an honest attempt to teach the readers in the most approachable way possible. In this process, combating any growing elitism in the AI research industry, Wu identifies in his personal experiences with another established researcher, Jelinek, the largest factor of success in academia: personal motivation. To this end, Wu works to foster interest before mathematical depth, all while maintaining the same rigor one would expect from a formal class.
I can only envy Rachel Wu, the translator of the English version of this book, and a fellow teacher of an introduction to ML class at MIT, for growing up with such a knowledgeable father.
The Beauty of Mathematics in Computer Science (BMCS) paints hope on a canvas which I have feared is but thinly veiling a research landscape saturated with papers consisting of throwing together integrations of pieces of exiting models or perhaps not much more than a simple increase in model size or parameter count without sufficiently supported insights. On such a bleak landscape, BMCS brings to life stories of researchers in natural language processing (NLP) and related fields with succinct but insightful explanations of the underlying mathematics. To the layman, "machine learning" (ML) is a foreign term, the all-consuming monster lurking in the corner where only those who have aced their high school and college math classes may approach. BMCS demonstrates that "machine learning" is no more than 10 sheets of book-sized paper filled with pictures, rightfully standing up against those who would rather use nebulous terms to show off their intelligence rather than attempting an honest explanation to a peer.
By no means is BMCS is meant to substitute a deeper dive into any of its topics. In Wu's struggle against his various NDAs, he makes an honest attempt to teach the readers in the most approachable way possible. In this process, combating any growing elitism in the AI research industry, Wu identifies in his personal experiences with another established researcher, Jelinek, the largest factor of success in academia: personal motivation. To this end, Wu works to foster interest before mathematical depth, all while maintaining the same rigor one would expect from a formal class.
I can only envy Rachel Wu, the translator of the English version of this book, and a fellow teacher of an introduction to ML class at MIT, for growing up with such a knowledgeable father.
March 2, 2020
这本书叫做数学之美，也许一部分原因是处于与吴军另外一本书《浪潮之巅》的书名呼应之故，但可能被更多的还是营销需要。
吴军以其从业经验和工作内容涉及到的工程领域的各种项目背景为主题，探讨其背后的数学运用、数学模型，偶尔还会穿插一些人物背景知识。比起书中频繁出现的数学公式，我更喜欢他讲述那些公式出现的背景。
阅读本书，需要一定的数学门槛，应用知识，或者专业兴趣，正是因为这些前提条件，如果没遇到对的人，反而会和数学之美的题设背道而驰。
吴军以其从业经验和工作内容涉及到的工程领域的各种项目背景为主题，探讨其背后的数学运用、数学模型，偶尔还会穿插一些人物背景知识。比起书中频繁出现的数学公式，我更喜欢他讲述那些公式出现的背景。
阅读本书，需要一定的数学门槛，应用知识，或者专业兴趣，正是因为这些前提条件，如果没遇到对的人，反而会和数学之美的题设背道而驰。
May 27, 2018
引句作者的話，世界上最好的學者總是能深入淺出的把大道理講給外行聽，而不是故弄玄虛地把簡單的問題複雜化。讀完這本書如沐春風，站在巨人肩膀上看IT巨頭如何利用數學這個工具來實現許多生活上待解決的問題，十分值得專家、業餘和外行讀的一本好書。
August 22, 2021
NLP、信息论等的入门读物，对于不懂数学的人也许会比较有帮助。
April 22, 2024
读不懂，放下了
July 5, 2016
冲着一端据说从本书中摘录的关于神经网络的形象讲解来看这本书，结果没有看到。不过书本身还是不错。
October 20, 2016
没想到专业书可读性也这么强。其实本书组织就适合没有基础的业外人士和很有基础的业内人士。加上例子和具体的数学公式，即浅显又专业。个人数学不好，第一遍读完就第二遍好好补补专业知识。
January 3, 2022
This book's Chinese title is the beauty of math, which confuses the reader a bit. Rather, this is like an introductory material to natural language processing, a subarea in AI. Good for beginners.
Displaying 1 - 12 of 12 reviews