Disentangled State Space Representations

Abstract

Sequential data often originates from diverse domains across which statistical regularities and domain specifics exist. To specifically learn cross-domain sequence representations, we introduce disentangled state space models (DSSM) -- a class of SSM in which domain-invariant state dynamics is explicitly disentangled from domain-specific information governing that dynamics. We analyze how such separation can improve knowledge transfer to new domains, and enable robust prediction, sequence manipulation and domain characterization. We furthermore propose an unsupervised VAE-based training procedure to implement DSSM in form of Bayesian filters. In our experiments, we applied VAE-DSSM framework to achieve competitive performance in online ODE system identification and regression across experimental settings, and controlled generation and prediction of bouncing ball video sequences across varying gravitational influences.

References in corpus (11)

• Stochastic Backpropagation and Approximate Inference in Deep Generative Models
• Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations
• A Recurrent Latent Variable Model for Sequential Data
• Decomposing Motion and Content for Natural Video Sequence Prediction
• Learning Stochastic Recurrent Networks
• Deep Kalman Filters
• Disentangled Sequential Autoencoder
• Deep Variational Bayes Filters: Unsupervised Learning of State Space Models from Raw Data
• Disentangling Factors of Variation Using Few Labels
• Deconfounding Reinforcement Learning in Observational Settings
• A General Method for Amortizing Variational Filtering

Cited by in corpus (7)

• On Disentangled Representations Learned From Correlated Data
• Learning Group Structure and Disentangled Representations of Dynamical Environments
• S2RMs: Spatially Structured Recurrent Modules
• The Differentiable Cross-Entropy Method
• Customizing Sequence Generation with Multi-Task Dynamical Systems
• Boxhead: A Dataset for Learning Hierarchical Representations
• Deep Variational Luenberger-type Observer for Stochastic Video Prediction