Disentangling Factors of Variation Using Few Labels

Abstract

Learning disentangled representations is considered a cornerstone problem in representation learning. Recently, Locatello et al. (2019) demonstrated that unsupervised disentanglement learning without inductive biases is theoretically impossible and that existing inductive biases and unsupervised methods do not allow to consistently learn disentangled representations. However, in many practical settings, one might have access to a limited amount of supervision, for example through manual labeling of (some) factors of variation in a few training examples. In this paper, we investigate the impact of such supervision on state-of-the-art disentanglement methods and perform a large scale study, training over 52000 models under well-defined and reproducible experimental conditions. We observe that a small number of labeled examples (0.01--0.5\% of the data set), with potentially imprecise and incomplete labels, is sufficient to perform model selection on state-of-the-art unsupervised models. Further, we investigate the benefit of incorporating supervision into the training process. Overall, we empirically validate that with little and imprecise supervision it is possible to reliably learn disentangled representations.

References in corpus (32)

• Auto-Encoding Variational Bayes
• Semi-Supervised Learning with Deep Generative Models
• Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations
• Deep Convolutional Inverse Graphics Network
• Disentangling by Factorising
• Recent Advances in Autoencoder-Based Representation Learning
• Visual Reinforcement Learning with Imagined Goals
• Towards a Definition of Disentangled Representations
• Weakly-supervised Disentangling with Recurrent Transformations for 3D View Synthesis
• Disentangling factors of variation in deep representations using adversarial training
• Unsupervised Feature Extraction by Time-Contrastive Learning and Nonlinear ICA
• Multi-Level Variational Autoencoder: Learning Disentangled Representations from Grouped Observations
• Discovering Hidden Factors of Variation in Deep Networks
• DARLA: Improving Zero-Shot Transfer in Reinforcement Learning
• On Causal and Anticausal Learning
• Learning Deep Disentangled Embeddings with the F-Statistic Loss
• A Disentangled Recognition and Nonlinear Dynamics Model for Unsupervised Learning
• Are Disentangled Representations Helpful for Abstract Visual Reasoning?
• Disentangled Sequential Autoencoder
• SCAN: Learning Hierarchical Compositional Visual Concepts
• SOM-VAE: Interpretable Discrete Representation Learning on Time Series
• Learning Latent Subspaces in Variational Autoencoders
• Curiosity Driven Exploration of Learned Disentangled Goal Spaces
• On the Transfer of Inductive Bias from Simulation to the Real World: a New Disentanglement Dataset
• Bayesian representation learning with oracle constraints
• Disentangling Disentanglement in Variational Autoencoders
• Learning the Irreducible Representations of Commutative Lie Groups
• Disentangling the independently controllable factors of variation by interacting with the world
• Understanding Visual Concepts with Continuation Learning
• Learning to Linearize Under Uncertainty
• Learning Disentangled Representations with Reference-Based Variational Autoencoders
• Competitive Training of Mixtures of Independent Deep Generative Models

Cited by in corpus (18)

• Are Disentangled Representations Helpful for Abstract Visual Reasoning?
• Low-Resource Knowledge-Grounded Dialogue Generation
• Zero-Resource Knowledge-Grounded Dialogue Generation
• Learning Causal Semantic Representation for Out-of-Distribution Prediction
• Learning Disentangled Behaviour Patterns for Wearable-based Human Activity Recognition
• Robustly Disentangled Causal Mechanisms: Validating Deep Representations for Interventional Robustness
• Desiderata for Representation Learning: A Causal Perspective
• Disentangled State Space Representations
• Disentanglement Learning for Variational Autoencoders Applied to Audio-Visual Speech Enhancement
• An Identifiable Double VAE For Disentangled Representations
• Learning and controlling the source-filter representation of speech with a variational autoencoder
• DynamicVAE: Decoupling Reconstruction Error and Disentangled Representation Learning
• An Improved Semi-Supervised VAE for Learning Disentangled Representations
• On Disentanglement in Gaussian Process Variational Autoencoders
• Odd-One-Out Representation Learning
• Robust Disentanglement of a Few Factors at a Time
• Deep Variational Luenberger-type Observer for Stochastic Video Prediction
• Controllable Recommenders using Deep Generative Models and Disentanglement