@
twistedmeadows @
zyxbcde 周志华教授的《机器学习》和 《 Deep Learning 》都特别不适合入门，读者对象要么是数学系物理系的想看看机器学习的理论，要么是你已经熟悉了基本的机器学习，想丰满下知识体系，需要一个导论的书。这些书的地位类似《 Java 编程思想》《算法导论》。
对于楼主，你最适合看的是吴恩达的
deeplearning.ai 的课，没有之一。这门课价格合适（免费收费版都有），吴恩达教授非常善于鼓励人，不是特别强调数学基础（虽然是必须的）。你如果认真看下去（记得做笔记），一定可以成功的。附上我的《
deeplearning.ai 课程学习总结》：
https://blog.wildcat.io/2018/02/summary-of-my-learning-experience-about-deeplearning-ai-zh/ 。希望可以对你有帮助。
另外对于楼上说自己看论文的，我相信你们也不会这么做的，希望不要瞎指点人。看论文是需要的，但一开始就看那简直是搞笑。
附上自己做的笔记：
Geoffrey Hinton 对学习深度学习有什么建议：
Read some literature but don ’ t read too much of it. For creative researchers, read something, find where you think most of people are doing wrong. I ’ m contrary in that sense. Just doesn ’ t feel right. Keep yourself. If you intuition is that, you should follow it and eventually it will be successful.
Replicate paper, finding the tricks that make it work.
Never stop programming. Good students: keep working on some work. Bad students: Give up easily.
General advice for AI starters: Basically read enough so you start developing intuitions, and trust your intuitions and go for it, and don ’ t be too worried if everybody else says it ’ s nonsense. There ’ s one thing, which is, if you think it ’ s a really good idea, and other people tell you it ’ s complete nonsense, then you know you ’ re really on to something. So one example of that is, when I first came up with variational methods, I sent email explaining to a former student of mine called Peter Brown, who knew a lot about EN. And he showed it to people who worked with him. And they said this guy ’ s drunk, or he ’ s just stupid. So they really, really thought it was nonsense. Now, it could have been partly the way I explained it, because I explained it in intuitive terms. But when you have what you think is a good idea, and other people think is complete rubbish, that ’ s the sign of a really good idea. （有点哲学 =。=）
One good piece of advice for new grad students is, see if you can find an advisor who has beliefs similar to yours. Because if you work on stuff that your advisor feels deeply about, you ’ ll get a lot of good advice and time from your advisor. If you work on stuff your advisor ’ s not interested in, all you ’ ll get is, you get some advice, but it won ’ t be nearly so useful.
How do you feel about people entering a PhD program, or in top team? A: 没有真正足够深刻理解这次革命的人，这不是第二次工业革命，但是已经很接近了。这改变了我们使用电脑的方式，然而现在的计算机科学部门是传统时代建立的。我以前的部门就不允许大量的人员搞这个方向，只让一小部分人去搞。
Geoffrey 过去老不理解 Andrew 老 push 他去做。虽然他老抱怨但是他做到了，并且现在认为这是个好想法，虽然工作量极大。
关于 AI 领域辩论自己的看法：Symbolic AI is a big mistake. 传统观念认为，words 进，words 出，中间思考的过程也应该是 类似 words 或者 strings 的东西。但是比如 CNN，是 pixels 入，pixels 出，但是中间是数学向量表示。他认为这才是正确的方向。
<amp-youtube data-videoid="-eyhCTvrEtE" layout="responsive" width="480" height="270"></amp-youtube>
事实胜于雄辩。