arXiv cs.AI· Jeffrey Jewett, William Solow, Sandhya Saisubramanian·· 10 小时前AI 评分29
OHCAM：用不确定性引导探索学习带条件与量化效果的动作模型
Learning Action Models with Conditional and Quantified Effects via Uncertainty-Guided Exploration
AI 导读
OHCAM 是一种在线动作模型学习方法，可从有限环境交互中学习带条件与量化效果的动作模型，通过最大化候选假设间的分歧主动选择信息量大的动作以降低不确定性，并对噪声观测保持鲁棒。该方法从简单假设起步、仅在数据不一致时扩展复杂条件以提升可扩展性，在六个基准规划域上样本效率高、解出的任务数显著多于基线，并在 Kinova Gen3 机器人上完成两项任务验证。
正文 · 原文
Abstract:Accurate action models are critical for effective planning. Existing action-model learning methods largely assume simple action representations or become computationally intractable when learning conditional and quantified effects. We present Online Hypothesis-Driven Conditional Action Model Learning (OHCAM), an online approach for learning action models with conditional and quantified effects from limited interactions with the environment. OHCAM maintains a belief over hypothesized action models and actively selects informative actions to reduce uncertainty by maximizing disagreement among competing hypotheses, while being robust to noisy observations. To enable scalability, OHCAM begins with a small set of simple action model hypotheses and expands to more complex conditions only when the current hypotheses become inconsistent with the data. Experiments on six benchmark planning domains demonstrate that OHCAM is sample efficient in learning action models that solve substantially more tasks than baselines, even with observation noise. We validate OHCAM on two tasks using a Kinova Gen3 robot, demonstrating the real-world applicability of our approach.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2608.30955 [cs.AI] |
| (or arXiv:2608.30955v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2608.30955 arXiv-issued DOI via DataCite |
Submission history
From: Jeffrey Jewett [view email]
[v1]
Mon, 31 Aug 2026 15:23:43 UTC (7,636 KB)
[v2]
Mon, 9 Nov 2026 20:04:24 UTC (7,638 KB)
来源：arXiv cs.AI · arxiv.org