Evaluating Model-Agnostic Meta-Learning on MetaWorld ML10 Benchmark: Fast Adaptation in Robotic Manipulation Tasks

TL;DR

This paper evaluates the effectiveness of Model-Agnostic Meta-Learning (MAML) combined with TRPO on the MetaWorld ML10 benchmark, demonstrating rapid adaptation in robotic manipulation tasks but also revealing generalization challenges.

Contribution

It provides a comprehensive analysis of MAML-TRPO's ability to enable few-shot learning across diverse robotic manipulation tasks, highlighting both successes and limitations.

Findings

MAML achieves 21.0% success on training tasks after one shot.

Test task performance plateaus, indicating a generalization gap.

High variance in adaptation success across different manipulation skills.

Abstract

Meta-learning algorithms enable rapid adaptation to new tasks with minimal data, a critical capability for real-world robotic systems. This paper evaluates Model-Agnostic Meta-Learning (MAML) combined with Trust Region Policy Optimization (TRPO) on the MetaWorld ML10 benchmark, a challenging suite of ten diverse robotic manipulation tasks. We implement and analyze MAML-TRPO's ability to learn a universal initialization that facilitates few-shot adaptation across semantically different manipulation behaviors including pushing, picking, and drawer manipulation. Our experiments demonstrate that MAML achieves effective one-shot adaptation with clear performance improvements after a single gradient update, reaching final success rates of 21.0% on training tasks and 13.2% on held-out test tasks. However, we observe a generalization gap that emerges during meta-training, where performance on test tasks plateaus while training task performance continues to improve. Task-level analysis reveals high variance in adaptation effectiveness, with success rates ranging from 0% to 80% across different manipulation skills. These findings highlight both the promise and current limitations of gradient-based meta-learning for diverse robotic manipulation, and suggest directions for future work in task-aware adaptation and structured policy architectures.