深度学习资源
简介
深度学习(Deep Learning)是机器学习(Machine Learning)的一个分支,一般指代“深度神经网络”(Deep Neural
Network)。历史上,人工神经网络(Artificial Neural
Networks)经历了三次发展浪潮:20世纪40年代到60年代,神经网络以“控制论”(cybernetics)闻名;20世纪80年代到90年代,表现为“联结主义”(connectionism);2006年至今,以“深度学习”之名复兴。得益于与日俱增的数据量和计算能力(GPU,
TPU),深度学习已经成功应用于计算机视觉、语音识别、自然语言处理、推荐系统等领域。
本组研究成果
- Jian Tang, Zhaoshi Meng, XuanLong Nguyen, Qiaozhu Mei, Ming Zhang. Understanding the Limiting Factors
of Topic Modeling via Posterior Contraction Analysis. ICML 2014: 190-198 (Best Paper)
- Jian Tang, Meng Qu, Mingzhe Wang, Ming Zhang, Jun Yan, Qiaozhu Mei. LINE: Large-scale
Information Network Embedding. WWW 2015: 1067-1077.
- Chenguang Wang, Yangqiu Song, Ahmed El-Kishky, Dan Roth, Ming Zhang, Jiawei Han. Incorporating World Knowledge
to Document Clustering via Heterogeneous Information Networks. KDD 2015: 1215-1224.
- Jian Tang, Jingzhou Liu, Ming Zhang, Qiaozhu Mei. Visualizing
Large-scale and High-dimensional Data. WWW 2016: 287-297. Best
paper nominee.
- Xiang Li, Lili Mou, RuiYan, Ming Zhang. StalemateBreaker: A Proactive Content-Introducing
Approach to Automatic Human-Computer Conversation. IJCAI 2016: 2845-2851.《每日邮报》报道;北大新闻报道
- Yin Yichun, Wei Furu, Dong Li, Xu Kaimeng, Zhang Ming and Zhou Ming. Unsupervised Word and Dependency
Path Embeddings for Aspect Term Extraction. IJCAI 2016: 2979-2985.
- Yiping Song, Cheng-Te Li, Jian-Yun Nie, Ming Zhang, Dongyan Zhao, Rui Yan. An Ensemble of Retrieval-Based and
Generation-Based Human-Computer Conversation Systems. IJCAI 2018: 4382-4388.
- Luchen Liu, Jianhao Shen, Ming Zhang, Zichang Wang, Jian Tang. Learning the Joint Representation of
Heterogeneous Temporal Events for Clinical Endpoint Prediction. AAAI 2018: 109-116.
基础教程
-
http://deeplearning.net/tutorial/
-
《神经网络与深度学习》讲义,邱锡鹏
-
《机器学习》完整版笔记,李宏毅
参考书籍
-
Deep Learning, Yoshua Bengio, Ian Goodfellow, Aaron
Courville, MIT Press, In preparation. 网页版
-
周志华《机器学习》
-
李航《统计学习方法》
-
Bishop et.al. 《Pattern Recognition and Machine Learning》
经典论文
- Hinton, Geoffrey E. “Deterministic Boltzmann learning performs steepest descent in
weight-space.” Neural computation 1.1 (1989): 143-150.
- Hinton, Geoffrey E., and Ruslan R. Salakhutdinov. “Reducing the dimensionality of data with
neural networks.” Science 313.5786 (2006): 504-507.
- LeCun, Yann, Yoshua Bengio, and Geoffrey Hinton. “Deep learning.” Nature 521, no. 7553(2015):
436-444.
- Yoshua Bengio, Aaron Courville, Pascal Vincent. “Representation Learning: A Review and New
Perspectives.” Arxiv, 2012.
- Le Roux, Nicolas, and Yoshua Bengio. “Deep belief networks are compact universal
approximators.” Neural computation 22.8 (2010): 2192-2207.
- Hochreiter, Sepp, and Jürgen Schmidhuber. “Long short-term memory.” Neural computation 9.8
(1997): 1735-1780.
- Alex Krizhevsky, Ilya Sutskever, Geoffrey E Hinton, “ImageNet Classification with Deep
Convolutional Neural Networks.” NIPS 2012.
- Mikolov, Tomas, Ilya Sutskever, Kai Chen, Greg S. Corrado, and Jeff Dean. “Distributed
representations of words and phrases and their compositionality.” In Advances in Neural
Information Processing Systems, pp. 3111-3119. 2013.
公开课
-
http://videolectures.net/deeplearning2015_montreal
-
http://videolectures.net/deeplearning2016_montreal
-
http://videolectures.net/kdd2014_salakhutdinov_deep_learning
框架(排名不分先后)
-
TensorFlow
官方网站:https://www.tensorflow.org/
TensorFlow是谷歌开源的机器学习框架,提供可靠的python和C++接口,Go和Java的接口仍在开发中。
-
Theano
官方网站:http://deeplearning.net/software/theano/
Theano是元老级的深度学习框架,提供python接口,具有良好的计算图抽象方式.
-
Torch
官方网站:http://torch.ch/
Torch是用Lua语言编写的计算框架,有很多已定义模型,容易编写自己的层类型并在GPU上运行,但不适合RNN。
-
Caffe
官方网站:http://caffe.berkeleyvision.org/
Caffe是应用较为广泛的机器视觉库,最适合于图像处理,不适合文本、声音等数据类型的深度学习应用.
-
CNTK
官方网站:https://github.com/Microsoft/CNTK
CNTK是微软开源的深度学习框架,主要包括前馈DNN、卷积网络和循环网络。
-
MXNet
官方网站:https://github.com/dmlc/mxnet
MXNet是陈天奇等开发的深度学习框架,目前已被亚马逊选中成为其深度学习平台。特点是运行速度快,内存管理效率高。
其它资源
-
http://deeplearning.net/reading-list/
-
https://github.com/ty4z2008/Qix/blob/master/dl.md
-
Andrew Ng 的机器学习网络公开课
-
Andrew Moore 的机器学习算法教程
-
无监督的特征学习和深度学习(中文翻译)
-
张志华公开课: 机器学习导论、
统计机器学习