Discovering Adaptable Symbolic Algorithms from Scratch

TL;DR

This paper introduces AutoRobotics-Zero, a novel AutoML-based method that evolves adaptable, interpretable control algorithms capable of handling sudden environmental changes in robotics, outperforming neural network baselines.

Contribution

We propose AutoRobotics-Zero, a method that evolves zero-shot adaptable control policies with full expressive power, enabling rapid adaptation and interpretability in robotic control tasks.

Findings

ARZ outperforms neural network baselines in robustness to environmental changes

ARZ produces simple, interpretable control policies

ARZ successfully adapts to limb failures in quadruped robots

Abstract

Autonomous robots deployed in the real world will need control policies that rapidly adapt to environmental changes. To this end, we propose AutoRobotics-Zero (ARZ), a method based on AutoML-Zero that discovers zero-shot adaptable policies from scratch. In contrast to neural network adaptation policies, where only model parameters are optimized, ARZ can build control algorithms with the full expressive power of a linear register machine. We evolve modular policies that tune their model parameters and alter their inference algorithm on-the-fly to adapt to sudden environmental changes. We demonstrate our method on a realistic simulated quadruped robot, for which we evolve safe control policies that avoid falling when individual limbs suddenly break. This is a challenging task in which two popular neural network baselines fail. Finally, we conduct a detailed analysis of our method on a novel and challenging non-stationary control task dubbed Cataclysmic Cartpole. Results confirm our findings that ARZ is significantly more robust to sudden environmental changes and can build simple, interpretable control policies.