Weakly Supervised Disentanglement with Guarantees

Learning disentangled representations that correspond to factors of variation in real-world data is critical to interpretable and human-controllable machine learning. Recently, concerns about the viability of learning disentangled representations in a purely unsupervised manner has spurred a shift toward the incorporation of weak supervision. However, there is currently no formalism that identifies when and how weak supervision will guarantee disentanglement. To address this issue, we provide a theoretical framework to assist in analyzing the disentanglement guarantees (or lack thereof) conferred by weak supervision when coupled with learning algorithms based on distribution matching. We empirically verify the guarantees and limitations of several weak supervision methods (restricted labeling, match-pairing, and rank-pairing), demonstrating the predictive power and usefulness of our theoretical framework.

[1]  Michael C. Mozer,et al.  Learning Deep Disentangled Embeddings with the F-Statistic Loss , 2018, NeurIPS.

[2]  Yang Song,et al.  Learning Fine-Grained Image Similarity with Deep Ranking , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[3]  Pascal Vincent,et al.  Representation Learning: A Review and New Perspectives , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Sebastian Nowozin,et al.  Multi-Level Variational Autoencoder: Learning Disentangled Representations from Grouped Observations , 2017, AAAI.

[5]  Bernhard Schölkopf,et al.  The Incomplete Rosetta Stone problem: Identifiability results for Multi-view Nonlinear ICA , 2019, UAI.

[6]  Christopher Burgess,et al.  beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework , 2016, ICLR 2016.

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

[8]  Christopher K. I. Williams,et al.  A Framework for the Quantitative Evaluation of Disentangled Representations , 2018, ICLR.

[9]  Andriy Mnih,et al.  Disentangling by Factorising , 2018, ICML.

[10]  Yedid Hoshen,et al.  Demystifying Inter-Class Disentanglement , 2020, ICLR.

[11]  Roger B. Grosse,et al.  Isolating Sources of Disentanglement in Variational Autoencoders , 2018, NeurIPS.

[12]  Yu-Ding Lu,et al.  DRIT++: Diverse Image-to-Image Translation via Disentangled Representations , 2020, International Journal of Computer Vision.

[13]  Kayhan Batmanghelich,et al.  Weakly Supervised Disentanglement by Pairwise Similarities , 2019, AAAI.

[14]  Abhishek Kumar,et al.  Variational Inference of Disentangled Latent Concepts from Unlabeled Observations , 2017, ICLR.

[15]  Richard S. Zemel,et al.  Learning Latent Subspaces in Variational Autoencoders , 2018, NeurIPS.

[16]  Heiga Zen,et al.  Sample Efficient Adaptive Text-to-Speech , 2018, ICLR.

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

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

[19]  Y. LeCun,et al.  Learning methods for generic object recognition with invariance to pose and lighting , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[20]  Lalana Kagal,et al.  Explaining Explanations: An Overview of Interpretability of Machine Learning , 2018, 2018 IEEE 5th International Conference on Data Science and Advanced Analytics (DSAA).

[21]  Stefan Bauer,et al.  Interventional Robustness of Deep Latent Variable Models , 2018, ArXiv.

[22]  Takeru Miyato,et al.  cGANs with Projection Discriminator , 2018, ICLR.

[23]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[24]  David Pfau,et al.  Towards a Definition of Disentangled Representations , 2018, ArXiv.

[25]  Dana H. Brooks,et al.  Structured Disentangled Representations , 2018, AISTATS.

[26]  Gert R. G. Lanckriet,et al.  Metric Learning to Rank , 2010, ICML.

[27]  Mykel J. Kochenderfer,et al.  Rethinking Style and Content Disentanglement in Variational Autoencoders , 2018, ICLR.

[28]  Yuting Zhang,et al.  Deep Visual Analogy-Making , 2015, NIPS.

[29]  Yedid Hoshen,et al.  Latent Optimization for Non-adversarial Representation Disentanglement , 2019, ArXiv.

[30]  Bernhard Schölkopf,et al.  Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations , 2018, ICML.