Hidden Parameter Recurrent State Space Models For Changing Dynamics Scenarios

TL;DR

The paper introduces HiP-RSSMs, a new framework for modeling changing dynamics in time series data using latent parameters, improving over traditional RSSMs in robotic benchmarks.

Contribution

It proposes a novel latent parameter approach for RSSMs that handles dynamic changes without variational inference, enhancing multi-task learning in control systems.

Findings

HiP-RSSMs outperform RSSMs on robotic benchmarks.

The model effectively captures changing dynamics.

It simplifies inference without approximations.

Abstract

Recurrent State-space models (RSSMs) are highly expressive models for learning patterns in time series data and system identification. However, these models assume that the dynamics are fixed and unchanging, which is rarely the case in real-world scenarios. Many control applications often exhibit tasks with similar but not identical dynamics which can be modeled as a latent variable. We introduce the Hidden Parameter Recurrent State Space Models (HiP-RSSMs), a framework that parametrizes a family of related dynamical systems with a low-dimensional set of latent factors. We present a simple and effective way of learning and performing inference over this Gaussian graphical model that avoids approximations like variational inference. We show that HiP-RSSMs outperforms RSSMs and competing multi-task models on several challenging robotic benchmarks both on real-world systems and simulations.