Learning STRIPS Action Models with Classical Planning
This addresses the challenge of automating action model acquisition in AI planning, which is incremental as it builds on existing compilation methods.
The paper tackles the problem of learning STRIPS action models from examples by compiling the inductive learning task into a classical planning task, achieving flexibility in handling varying input knowledge such as plans or just initial and final states.
This paper presents a novel approach for learning STRIPS action models from examples that compiles this inductive learning task into a classical planning task. Interestingly, the compilation approach is flexible to different amounts of available input knowledge; the learning examples can range from a set of plans (with their corresponding initial and final states) to just a pair of initial and final states (no intermediate action or state is given). Moreover, the compilation accepts partially specified action models and it can be used to validate whether the observation of a plan execution follows a given STRIPS action model, even if this model is not fully specified.