用少量专家数据实现跨域策略迁移，稳定高效。
Semi-Supervised Cross-Domain Imitation Learning
- 结合少量目标域专家数据与无标签轨迹，设计半监督跨域模仿学习方法。
- 在MuJoCo和Robosuite上性能超越基线，提升稳定性和数据效率。
- 适合缺乏大量标注数据的机器人控制场景，尤其关注低监督学习。
跨域模仿学习（CDIL）通过在不同领域间迁移专家知识，加速策略学习，适用于专家数据收集成本高的场景。现有方法要么依赖代理任务和显式对齐的监督方法，要么是无需成对数据的无监督方法，但常不稳定。本文提出半监督跨域模仿学习（SS-CDIL）新设定，并首次给出具有理论支持的算法。该方法仅使用离线数据，包括少量目标域专家示范和若干无标签的非完美轨迹。为缓解域差异，我们设计了一种新型跨域损失函数以学习域间状态-动作映射，并引入自适应权重函数平衡源域与目标域知识。在MuJoCo和Robosuite上的实验表明，相比基线方法，本方法持续取得性能提升，验证了其在最小监督条件下实现稳定且数据高效的策略学习能力。代码已开源：https://github.com/NYCU-RL-Bandits-Lab/CDIL。
原文摘要 · Abstract (English)
Cross-domain imitation learning (CDIL) accelerates policy learning by transferring expert knowledge across domains, which is valuable in applications where the collection of expert data is costly. Existing methods are either supervised, relying on proxy tasks and explicit alignment, or unsupervised, aligning distributions without paired data, but often unstable. We introduce the Semi-Supervised CDIL (SS-CDIL) setting and propose the first algorithm for SS-CDIL with theoretical justification. Our method uses only offline data, including a small number of target expert demonstrations and some unlabeled imperfect trajectories. To handle domain discrepancy, we propose a novel cross-domain loss function for learning inter-domain state-action mappings and design an adaptive weight function to balance the source and target knowledge. Experiments on MuJoCo and Robosuite show consistent gains over the baselines, demonstrating that our approach achieves stable and data-efficient policy learning with minimal supervision. Our code is available at~ https://github.com/NYCU-RL-Bandits-Lab/CDIL.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。