Searching for Planning Operators with Context-Dependent and Probabilistic Effects

Abstract

Providing a complete and accurate domain model for an agent situated in a complex environmentcan be an extremely difficult task. Actions mayhave different effects depending on the context in which they are taken, and actions mayormay not induce their intended effects, with the probability of success again depending on context. We present an algorithm for automatically learning planning operators with context-dependent and probabilistic effects in environments where exogenous events change the state of the world. Empirical results show that the algorithm successfully finds operators that capture the true structure of an agent's interactions with its environment, and avoids spurious associations between actions and exogenous events.