强化学习
Reinforcement Learning
Richard S. Sutton · 初版 1998 年
18人推荐
30位读者提及
34次提及
为什么值得读
- 由领域先驱撰写，内容详实且极易入门
- 前言符号表清晰标注非初等代数标记
- 深入阐释强化学习与马尔可夫决策过程的联系
Reinforcement learning 是指学习一种从情境到动作的映射，从而最大化标量奖励或强化信号。与大多数形式的 machine learning 不同，学习者不会被告知该采取何种动作，而是必须通过不断尝试来发现哪些动作能带来最高回报。在最有趣且最具挑战性的情况下，动作不仅会影响即时奖励，还会影响下一个情境，并进而影响随后的所有奖励。这两大特征——trial-and-error search 与 delayed reward——正是 reinforcement learning 最重要的本质特征……
Hacker News 上怎么说
正在加载完整评论……
……毫无疑问，这些是该领域最顶级的教材：- [0] Pattern Recognition and Machine Learning (Information Science and Statistics)，另外还有：- [1] The Elements of Statistical Learning - [2] Barto 和 Sutton 的《Reinforcement Learning: An Introduction》 - [3] Aaron Courville、Ian Goodfellow 和 Yoshua Bengio 的《Deep Learning》 - [4] Neural Network Methods for Natural Language Processing (Synthesis Lectures on…
…hands down these are some of the best related textbooks: - [0] Pattern Recognition and Machine Learning (Information Science and Statistics) and also: - [1] The Elements of Statistical Learning - [2] Reinforcement Learning: An Introduction by Barto and Sutton - [3] The Deep Learning by Aaron Courville, Ian Goodfellow, and Yoshua Bengio - [4] Neural Network Methods for Natural Language Processing (Synthesis Lectures on…
……Multilevel/Hierarchical Models (Gelman & Hill): https://www.amazon.ca/gp/product/052168689X/ - Statistical and Data Analysis for Financial Engineering (Ruppert & Matteson): https://www.amazon.ca/gp/product/1493926136/ - Reinforcement Learning (Sutton & Barto): https://www.amazon.ca/Reinforcement-Learning-Introduction-Ri... - Deep Learning (Godfellow, Bengio & Courville): https://www.amazon.ca/Deep-Learning-Ian-Goodfellow/dp/026203... - Networks…
…Multilevel/Hierarchical Models (Gelman & Hill): https://www.amazon.ca/gp/product/052168689X/ - Statistical and Data Analysis for Financial Engineering (Ruppert & Matteson): https://www.amazon.ca/gp/product/1493926136/ - Reinforcement Learning (Sutton & Barto): https://www.amazon.ca/Reinforcement-Learning-Introduction-Ri... - Deep Learning (Godfellow, Bengio & Courville): https://www.amazon.ca/Deep-Learning-Ian-Goodfellow/dp/026203... - Networks…
……超越了以往所有的电脑西洋双陆棋程序。在使用 80 个隐藏单元进行了约 150 万局自对弈（self-play）后，它的水平停止了提升。顺便告诉其他人：时序差分（TD）的先驱之一 Richard Sutton 把他的教材《Reinforcement Learning: An Introduction》免费公开在了他的网站上：http://incompleteideas.net/sutton/（MIT Press 也提供了链接）
…surpassed all previous computer backgammon programs. It stopped improving after about 1,500,000 games (self-play) using 80 hidden units. For others: Richard Sutton, one of the pioneers of TD makes his Reinforcement Learning: An Introduction textbook available for free on his website: http://incompleteideas.net/sutton/ (MIT Press also links to it)
不知道作者能不能看到这条，但我求求你们了，请务必在前言里放一个符号表。Sutton 和 Barto 的那本《强化学习》对基本上所有非初等代数的符号都做了标注，这太有用了。光是把指示函数（indicator function）这类我以前没见过的符号标出来，就极其宝贵。尤其是……
Not sure if the authors will read this or not but I beg of you, please put a table of notation in the forward. The Sutton and Barto Reinforcement Learning book did that for basically every notation that wasn't basic algebra and it's been extremely helpful. Just labeling things I had never seen before, like indicator functions, was extremely valuable. Especially…
可以考虑读一下 Rich Sutton 的《Reinforcement Learning: An Introduction》[0]。它非常通俗易懂（而且可能是该领域使用最广泛的教材），并且深入讲解了 RL 和 MDP 之间的联系。如果你想要一个宏观的……
Consider reading "Reinforcement Learning, an Introduction" by Rich Sutton[0]. It is very accessible (and probably the most used textbook in the field), and goes into detail about the connection between RL and MDPs. If you want a high-level…
正如其他人所说，这似乎是强化学习问题的一个泛化。如果想对这一主题有一个很好的入门了解，可以看看 Sutton 和 Barto 写的《Reinforcement Learning》[1]。读完前几章后，你就能看懂那个公式的大部分内容了。[1] http://webdocs.cs.ualberta.ca/~sutton/book/the-book.html
Like others have said, this seems to be a generalization of the problem of reinforcement learning. For a good introduction to the subject, check out Reinforcement Learning by Sutton & Barto [1]. After reading the first few chapters, you'll be able to understand most of that equation. [1] http://webdocs.cs.ualberta.ca/~sutton/book/the-book.html