BayesSimIG: Scalable Parameter Inference for Adaptive Domain Randomization with IsaacGym

TL;DR

BayesSimIG combines likelihood-free inference with GPU-accelerated simulation in IsaacGym, enabling scalable, high-dimensional parameter inference for complex robotics tasks with real-time visualization.

Contribution

It introduces a GPU-accelerated library integrating BayesSim with IsaacGym, supporting large-scale, high-dimensional parameter inference for reinforcement learning.

Findings

Supports over 10,000 parallel environments

Enables inference with more than 100 parameters

Provides visualization of high-dimensional posteriors

Abstract

BayesSim is a statistical technique for domain randomization in reinforcement learning based on likelihood-free inference of simulation parameters. This paper outlines BayesSimIG: a library that provides an implementation of BayesSim integrated with the recently released NVIDIA IsaacGym. This combination allows large-scale parameter inference with end-to-end GPU acceleration. Both inference and simulation get GPU speedup, with support for running more than 10K parallel simulation environments for complex robotics tasks that can have more than 100 simulation parameters to estimate. BayesSimIG provides an integration with TensorBoard to easily visualize slices of high-dimensional posteriors. The library is built in a modular way to support research experiments with novel ways to collect and process the trajectories from the parallel IsaacGym environments.