English(EN) Difference-Aware Retrieval Policies for Imitation Learning

新的DARP方法增强了模仿学习的泛化能力

作者 PulseAugur 编辑部 · [2 个来源] · 2026-06-08 17:18

研究人员开发了一种名为差异感知检索策略（DARP）的新型模仿学习方法。该方法通过在推理过程中使用训练数据来提高泛化能力，并根据查询状态的k个最近邻及其相对距离来预测动作。DARP在机器人技术和连续控制等多个领域取得了比标准行为克隆显著的性能提升。 AI

影响增强了模仿学习的泛化能力，有望改进机器人控制和自主系统。

排序理由该集群包含一篇详细介绍模仿学习新方法的学术论文。

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv cs.AI TIER_1 English(EN) · Quinn Pfeifer, Ethan Pronovost, Paarth Shah, Khimya Khetarpal, Siddhartha Srinivasa, Abhishek Gupta · 2026-06-09 04:00

面向模仿学习的差异感知检索策略

arXiv:2606.09758v1 Announce Type: cross Abstract: Parametric imitation learning via behavior cloning can suffer from poor generalization to out-of-distribution states due to compounding errors during deployment. We show that reusing the training data during inference via a semi-p…
arXiv cs.AI TIER_1 English(EN) · Abhishek Gupta · 2026-06-08 17:18

用于模仿学习的差异感知检索策略

Parametric imitation learning via behavior cloning can suffer from poor generalization to out-of-distribution states due to compounding errors during deployment. We show that reusing the training data during inference via a semi-parametric retrieval-based imitation learning appro…