Heuristic Adaptation of Potentially Misspecified Domain Support for Likelihood-Free Inference in Stochastic Dynamical Systems

TL;DR

This paper introduces three heuristic methods to adapt support in likelihood-free inference, improving parameter estimation and policy robustness in stochastic dynamical systems, especially in robotics applications.

Contribution

The paper proposes three novel heuristic variants (EDGE, MODE, CENTRE) for support adaptation in likelihood-free inference, addressing support misspecification issues.

Findings

Heuristic support adaptation improves parameter inference accuracy.

Enhanced robustness in policy learning for DLO manipulation.

Support adaptation leads to finer classification of DLO properties.

Abstract

In robotics, likelihood-free inference (LFI) can provide the domain distribution that adapts a learnt agent in a parametric set of deployment conditions. LFI assumes an arbitrary support for sampling, which remains constant as the initial generic prior is iteratively refined to more descriptive posteriors. However, a potentially misspecified support can lead to suboptimal, yet falsely certain, posteriors. To address this issue, we propose three heuristic LFI variants: EDGE, MODE, and CENTRE. Each interprets the posterior mode shift over inference steps in its own way and, when integrated into an LFI step, adapts the support alongside posterior inference. We first expose the support misspecification issue and evaluate our heuristics using stochastic dynamical benchmarks. We then evaluate the impact of heuristic support adaptation on parameter inference and policy learning for a dynamic deformable linear object (DLO) manipulation task. Inference results in a finer length and stiffness classification for a parametric set of DLOs. When the resulting posteriors are used as domain distributions for sim-based policy learning, they lead to more robust object-centric agent performance.