Learning Action Models with Conditional and Quantified Effects via Uncertainty-Guided Exploration
2026-08-31 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors introduce OHCAM, a method that helps computers learn how actions work in different situations by testing and updating their guesses based on what they observe. Unlike previous methods, OHCAM can handle complicated actions that depend on certain conditions and counts, and it works even when observations are noisy or imperfect. It starts with simple guesses and only gets more complex if needed, making it efficient in learning from fewer examples. The authors tested OHCAM in simulations and on a real robot, showing it can learn effective action models that help solve tasks reliably.
action modelplanningconditional effectsquantified effectsonline learninghypothesis testingnoisy observationssample efficiencyrobot manipulationKinova Gen3
Authors
Jeffrey Jewett, William Solow, Sandhya Saisubramanian
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.