search

ML-HK / paper-discussion-group

Discussion group of machine learning papers in HKUST
9 stars 0 forks source link

[Paper Discussion] 5th session #6

Closed sxjscience closed 8 years ago

sxjscience commented 8 years ago

The 5th session will be held on 2016/10/28. Voting period for this session ends up on 2016/10/23. We can vote/recommend papers in #3

sodabeta7 commented 8 years ago

I can give a brief introduction on paper Harnessing Deep Neural Networks with Logic Rules for this session.

sxjscience commented 8 years ago

Due to typhoon, we will postpone this session to 2016/11/04.

sxjscience commented 8 years ago

Paper for this session:

By @yyuanad

InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets (Xi Chen, Yan Duan, Rein Houthooft, John Schulman, Ilya Sutskever, Pieter Abbeel) (NIPS2016) This paper describes InfoGAN, an information-theoretic extension to the Generative Adversarial Network that is able to learn disentangled representations in a completely unsupervised manner. InfoGAN is a generative adversarial network that also maximizes the mutual information between a small subset of the latent variables and the observation. We derive a lower bound of the mutual information objective that can be optimized efficiently. Specifically, InfoGAN successfully disentangles writing styles from digit shapes on the MNIST dataset, pose from lighting of 3D rendered images, and background digits from the central digit on the SVHN dataset. It also discovers visual concepts that include hair styles, presence/absence of eyeglasses, and emotions on the CelebA face dataset. Experiments show that InfoGAN learns interpretable representations that are competitive with representations learned by existing supervised methods.