技能组合中学习技能的经验引导式初始配置搜索

中文摘要 AI 辅助

在新环境中部署冻结的学习技能（如视觉-语言-行动（VLA）策略）需要确定能支持可靠执行的初始配置。在技能组合中，初始配置不仅影响当前技能，还会传递给后续技能的物理状态，因此单个技能的成功执行并不一定意味着组合任务能完成。通过大量滚动估计目标特定能力在实际部署中成本高昂，而在环境变化下直接复用历史经验可能不可靠。我们提出EVIS，一种经验引导且行为验证的初始配置搜索框架，用于在有限的目标交互下发现可靠的初始配置。EVIS利用历史执行经验对有前景的候选进行排序，并通过目标环境的行为验证它们是否仍然有效。我们在使用冻结VLA策略的单技能和两阶段操作任务上评估EVIS，结果显示EVIS减少了平均目标环境查询次数，并在小交互预算下提高了可靠候选的发现效率。这些结果表明，结合历史引导与目标侧行为验证可降低在新环境中部署冻结学习技能的交互成本。

英文摘要

Deploying frozen learned skills, such as Vision-Language-Action (VLA) policies, in new environments requires identifying initiation configurations that support reliable execution. In skill composition, an initiation configuration affects not only the current skill but also the physical state passed to subsequent skills, so successful execution of an individual skill does not necessarily imply successful completion of the composed task. Estimating target-specific capability through extensive rollouts is costly in real-world deployment, while directly reusing historical experience can be unreliable under environment changes. We propose EVIS, an Experience-Guided and Behavior-Validated Initiation Search framework for discovering reliable initiation configurations under limited target interaction. EVIS uses historical execution experience to prioritize promising candidates and target-environment behavior to validate whether they remain effective. We evaluate EVIS on single-skill and two-stage manipulation tasks with frozen VLA policies. EVIS reduces mean target-environment queries and improves reliable candidate discovery under small interaction budgets. These results show that combining historical guidance with target-side behavioral validation can reduce the interaction cost of deploying frozen learned skills in new environments.

发表机构

• Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院）
• School of Electronic Science and Engineering, Nanjing University（南京大学电子科学与工程学院）
• Institute for Embodied Intelligence and Robotics, Tsinghua University（清华大学具身智能与机器人研究院）

机构由 AI 辅助整理，请以论文原文为准。