Google办的Deep Learning公开课

Udacity: deep-learning--ud730

By: Vincent Vanhoucke is a Principal Research Scientist at Google, working with the Google Brain team on deep learning research and infrastructure. He completed his Ph.D. at Stanford University on speech recognition, and now focusses his research on image and video understanding as well as mobile and robotic perception.

jxlin 发表于 2016-2-5 12:15
請問，在哪可以看到 hugo larochelle  的  deep learning 課？

他的youtube会有。他是sherbrooke 大学的教授。bengio和hinton的学生。 深度最根正苗红的。你去他的youtube和sherbrooke大学主页都能找到。

d1987115w 发表于 2016-2-2 16:06
请问能否推荐些深度学习的资料？多谢！

实际上楼主不一定是要执着于深度的，深度对于图像与语言等发杂有良好效果，但是实际上ML里面更多的人会去研究其他方向，不过学习的话，强烈推荐HUGO LAROCHELLE教授的deep learning的课程，然后DEEP LEARNING TUTORIAL 的theano代码希望能够读完并自己试着调试一次，theano细节又底层的设计虽然在debug和对于初学者的使用上有一点困难，但是绝对是对深度模型最直接的symbolic框架，然后Montreal Deep Learning Summer有机会可以申请试试。框架都是不必要的，自己喜欢用什么是什么。

实际上没有必要一味的去学习tensorflow跟风，tensorflow强大的SCALABLITY在课程和目前的开源框架中完全体现不出来。
课程对于深度学习更是走马冠花，对于初学者真的没有很大的价值，框架只要适合自己就好。这一点上单点的theano底层实现不比tensorflow学起来收获少。

http://www.udacity.com/course/deep-learning--ud730
Prerequisites and Requirements

This is an intermediate to advanced level course. Prior to taking this course, and in addition to the prerequisites and requirements outlined for the Machine Learning Engineer Nanodegree program, you should possess the following experience and skills:

Minimum 2 years of programming experience (preferably in Python)
Git and GitHub experience (assignment code is in a GitHub repo)
Basic machine learning knowledge (especially supervised learning)
Basic statistics knowledge (mean, variance, standard deviation, etc.)
Linear algebra (vectors, matrices, etc.)
Calculus (differentiation, integration, partial derivatives, etc.)




Lesson 1: From Machine Learning to Deep Learning

Understand the historical context and motivation for Deep Learning.
Set up a basic supervised classification task and train a black box classifier on it.
Train a logistic classifier “by hand”Optimize a logistic classifier using gradient descent, SGD, Momentum and AdaGrad.
Lesson 2: Deep Neural Networks

Train a simple deep network.
Effectively regularize a simple deep network.
Train a competitive deep network via model exploration and hyperparameter tuning.
Lesson 3: Convolutional Neural Networks

Train a simple convolutional neural net.
Explore the design space for convolutional nets.
Lesson 4: Deep Models for Text and Sequences

Train a text embedding model.
Train a LSTM model.

这可是个好东西，提醒一下课程中需要用到Tensorflow，which needs a Linux or Mac environment.

好！一直用ubuntu。。。

gyzjay 发表于 2016-2-2 12:56
实际上没有必要一味的去学习tensorflow跟风，tensorflow强大的SCALABLITY在课程和目前的开源框架中完全体现 ...

请问能否推荐些深度学习的资料？多谢！

gyzjay 发表于 2016-2-2 16:44
实际上楼主不一定是要执着于深度的，深度对于图像与语言等发杂有良好效果，但是实际上ML里面更多的人会去 ...

請問，在哪可以看到 hugo larochelle  的  deep learning 課？

这个课不值得上，基本是给tensorflow打广告的