Context is Everything: Implicit Identification for Dynamics Adaptation

TL;DR

The paper introduces IIDA, a method enabling predictive models to adapt to changing environment dynamics by implicitly inferring environment properties from limited data, improving performance in unseen scenarios.

Contribution

IIDA is a novel approach that does not require explicit knowledge of environment variations, allowing better adaptation in non-stationary settings.

Findings

IIDA reduces model error significantly.

IIDA achieves higher task performance.

Effective in both simulated and real robot tasks.

Abstract

Understanding environment dynamics is necessary for robots to act safely and optimally in the world. In realistic scenarios, dynamics are non-stationary and the causal variables such as environment parameters cannot necessarily be precisely measured or inferred, even during training. We propose Implicit Identification for Dynamics Adaptation (IIDA), a simple method to allow predictive models to adapt to changing environment dynamics. IIDA assumes no access to the true variations in the world and instead implicitly infers properties of the environment from a small amount of contextual data. We demonstrate IIDA's ability to perform well in unseen environments through a suite of simulated experiments on MuJoCo environments and a real robot dynamic sliding task. In general, IIDA significantly reduces model error and results in higher task performance over commonly used methods. Our code and robot videos are at https://bennevans.github.io/iida/