Guiding InfoGAN with Semi-supervision

In this paper we propose a new semi-supervised GAN architecture (ss-InfoGAN) for image synthesis that leverages information from few labels (as little as 0.22%, max. 10% of the dataset) to learn semantically meaningful and controllable data representations where latent variables correspond to label categories. The architecture builds on Information Maximizing Generative Adversarial Networks (InfoGAN) and is shown to learn both continuous and categorical codes and achieves higher quality of synthetic samples compared to fully unsupervised settings. Furthermore, we show that using small amounts of labeled data speeds-up training convergence. The architecture maintains the ability to disentangle latent variables for which no labels are available. Finally, we contribute an information-theoretic reasoning on how introducing semi-supervision increases mutual information between synthetic and real data.

[1]  Max Welling,et al.  Improved Variational Inference with Inverse Autoregressive Flow , 2016, NIPS 2016.

[2]  Max Welling,et al.  Auto-Encoding Variational Bayes , 2013, ICLR.

[3]  Xiaogang Wang,et al.  Deep Learning Face Attributes in the Wild , 2014, 2015 IEEE International Conference on Computer Vision (ICCV).

[4]  Max Welling,et al.  Semi-supervised Learning with Deep Generative Models , 2014, NIPS.

[5]  Wojciech Zaremba,et al.  Improved Techniques for Training GANs , 2016, NIPS.

[6]  Yoshua Bengio,et al.  Disentangling Factors of Variation via Generative Entangling , 2012, ArXiv.

[7]  Aaron C. Courville,et al.  Discriminative Regularization for Generative Models , 2016, ArXiv.

[8]  Andrew Y. Ng,et al.  Reading Digits in Natural Images with Unsupervised Feature Learning , 2011 .

[9]  Clément Farabet,et al.  Torch7: A Matlab-like Environment for Machine Learning , 2011, NIPS 2011.

[10]  Navdeep Jaitly,et al.  Adversarial Autoencoders , 2015, ArXiv.

[11]  Jonathon Shlens,et al.  Conditional Image Synthesis with Auxiliary Classifier GANs , 2016, ICML.

[12]  Frank D. Wood,et al.  Learning Disentangled Representations with Semi-Supervised Deep Generative Models , 2017, NIPS.

[13]  Lucas Theis,et al.  Amortised MAP Inference for Image Super-resolution , 2016, ICLR.

[14]  Joshua B. Tenenbaum,et al.  Separating Style and Content with Bilinear Models , 2000, Neural Computation.

[15]  Yuting Zhang,et al.  Learning to Disentangle Factors of Variation with Manifold Interaction , 2014, ICML.

[16]  David Barber,et al.  The IM algorithm: a variational approach to Information Maximization , 2003, NIPS 2003.

[17]  Yoshua Bengio,et al.  Generative Adversarial Networks , 2014, ArXiv.

[18]  Nikos Komodakis,et al.  Wide Residual Networks , 2016, BMVC.

[19]  Pieter Abbeel,et al.  InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets , 2016, NIPS.

[20]  Simon Osindero,et al.  Conditional Generative Adversarial Nets , 2014, ArXiv.

[21]  Ole Winther,et al.  Ladder Variational Autoencoders , 2016, NIPS.

[22]  Alex Krizhevsky,et al.  Learning Multiple Layers of Features from Tiny Images , 2009 .

[23]  Soumith Chintala,et al.  Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks , 2015, ICLR.

[24]  Xiaogang Wang,et al.  Deep Learning Multi-View Representation for Face Recognition , 2014, ArXiv.

[25]  Ruslan Salakhutdinov,et al.  Importance Weighted Autoencoders , 2015, ICLR.

[26]  Yann LeCun,et al.  Disentangling factors of variation in deep representation using adversarial training , 2016, NIPS.

[27]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[28]  David Pfau,et al.  Unrolled Generative Adversarial Networks , 2016, ICLR.

[29]  Joshua B. Tenenbaum,et al.  Deep Convolutional Inverse Graphics Network , 2015, NIPS.