Disentangling Controllable and Uncontrollable Factors of Variation by Interacting with the World

Abstract

We introduce a method to disentangle controllable and uncontrollable factors of variation by interacting with the world. Disentanglement leads to good representations and is important when applying deep neural networks (DNNs) in fields where explanations are required. This study attempts to improve an existing reinforcement learning (RL) approach to disentangle controllable and uncontrollable factors of variation, because the method lacks a mechanism to represent uncontrollable obstacles. To address this problem, we train two DNNs simultaneously: one that represents the controllable object and another that represents uncontrollable obstacles. For stable training, we applied a pretraining approach using a model robust against uncontrollable obstacles. Simulation experiments demonstrate that the proposed model can disentangle independently controllable and uncontrollable factors without annotated data.

arXiv commentRevised version

References in corpus (15)

• Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning?
• Towards A Rigorous Science of Interpretable Machine Learning
• Deep Convolutional Inverse Graphics Network
• Learning Disentangled Representations with Semi-Supervised Deep Generative Models
• Isolating Sources of Disentanglement in Variational Autoencoders
• InfoGAIL: Interpretable Imitation Learning from Visual Demonstrations
• The Consciousness Prior
• Learning Disentangled Joint Continuous and Discrete Representations
• Independently Controllable Factors
• Variational auto-encoding of protein sequences
• Path Integral Networks: End-to-End Differentiable Optimal Control
• Independently Controllable Features
• Challenges in Disentangling Independent Factors of Variation
• Reconstruction-Based Disentanglement for Pose-invariant Face Recognition
• Convolutional neural networks for structured omics: OmicsCNN and the OmicsConv layer

Cited by in corpus (4)

• A survey on intrinsic motivation in reinforcement learning
• Contingency-Aware Exploration in Reinforcement Learning
• PlaNet of the Bayesians: Reconsidering and Improving Deep Planning Network by Incorporating Bayesian Inference
• Scaling All-Goals Updates in Reinforcement Learning Using Convolutional Neural Networks