MetaWorld ML-10

The MetaWorld ML-10 benchmark is a suite of ten diverse robotic manipulation tasks used to evaluate the effectiveness of meta-learning algorithms, particularly in their ability to enable rapid adaptation to new tasks with minimal data.

Papers using MetaWorld ML-10 (2)

Context Representation via Action-Free Transformer encoder-decoder for Meta Reinforcement Learning2025Evaluating Model-Agnostic Meta-Learning on MetaWorld ML10 Benchmark: Fast Adaptation in Robotic Manipulation Tasks2025