A System for Morphology-Task Generalization via Unified Representation and Behavior Distillation

TL;DR

This paper introduces a unified graph-based representation and Transformer architecture for training a single policy capable of handling diverse agent morphologies and tasks, demonstrating improved multi-task performance and transfer capabilities.

Contribution

It proposes a morphology-task graph and a large-scale benchmark, MxT-Bench, enabling efficient multi-task learning and zero-shot transfer in morphology-task generalization.

Findings

Graph representation improves multi-task performance

Transformer architecture outperforms baselines

Enhanced zero-shot transfer and sample efficiency

Abstract

The rise of generalist large-scale models in natural language and vision has made us expect that a massive data-driven approach could achieve broader generalization in other domains such as continuous control. In this work, we explore a method for learning a single policy that manipulates various forms of agents to solve various tasks by distilling a large amount of proficient behavioral data. In order to align input-output (IO) interface among multiple tasks and diverse agent morphologies while preserving essential 3D geometric relations, we introduce morphology-task graph, which treats observations, actions and goals/task in a unified graph representation. We also develop MxT-Bench for fast large-scale behavior generation, which supports procedural generation of diverse morphology-task combinations with a minimal blueprint and hardware-accelerated simulator. Through efficient representation and architecture selection on MxT-Bench, we find out that a morphology-task graph representation coupled with Transformer architecture improves the multi-task performances compared to other baselines including recent discrete tokenization, and provides better prior knowledge for zero-shot transfer or sample efficiency in downstream multi-task imitation learning. Our work suggests large diverse offline datasets, unified IO representation, and policy representation and architecture selection through supervised learning form a promising approach for studying and advancing morphology-task generalization.