X-IL: Exploring the Design Space of Imitation Learning Policies

TL;DR

X-IL is a modular framework that systematically explores the design space of imitation learning policies, leading to the discovery of novel configurations with improved performance on robot learning benchmarks.

Contribution

The paper introduces X-IL, an open-source, modular framework enabling comprehensive exploration of IL policy design choices, facilitating better performance and insights.

Findings

Novel policy configurations outperform existing methods

Significant performance improvements on robot benchmarks

Insights into the strengths and weaknesses of design choices

Abstract

Designing modern imitation learning (IL) policies requires making numerous decisions, including the selection of feature encoding, architecture, policy representation, and more. As the field rapidly advances, the range of available options continues to grow, creating a vast and largely unexplored design space for IL policies. In this work, we present X-IL, an accessible open-source framework designed to systematically explore this design space. The framework's modular design enables seamless swapping of policy components, such as backbones (e.g., Transformer, Mamba, xLSTM) and policy optimization techniques (e.g., Score-matching, Flow-matching). This flexibility facilitates comprehensive experimentation and has led to the discovery of novel policy configurations that outperform existing methods on recent robot learning benchmarks. Our experiments demonstrate not only significant performance gains but also provide valuable insights into the strengths and weaknesses of various design choices. This study serves as both a practical reference for practitioners and a foundation for guiding future research in imitation learning.