GRAIL: Goal Recognition Alignment through Imitation Learning
This work addresses the challenge of aligning AI systems with human intentions by improving goal recognition in uncertain environments, representing an incremental advancement over existing methods.
The paper tackles the problem of accurately recognizing an agent's goals from its behavior by introducing GRAIL, which uses imitation learning and inverse reinforcement learning to learn goal-directed policies from potentially suboptimal demonstrations, resulting in F1-score improvements of up to 0.5 under biased behavior and 0.1-0.4 under suboptimal or noisy conditions.
Understanding an agent's goals from its behavior is fundamental to aligning AI systems with human intentions. Existing goal recognition methods typically rely on an optimal goal-oriented policy representation, which may differ from the actor's true behavior and hinder the accurate recognition of their goal. To address this gap, this paper introduces Goal Recognition Alignment through Imitation Learning (GRAIL), which leverages imitation learning and inverse reinforcement learning to learn one goal-directed policy for each candidate goal directly from (potentially suboptimal) demonstration trajectories. By scoring an observed partial trajectory with each learned goal-directed policy in a single forward pass, GRAIL retains the one-shot inference capability of classical goal recognition while leveraging learned policies that can capture suboptimal and systematically biased behavior. Across the evaluated domains, GRAIL increases the F1-score by more than 0.5 under systematically biased optimal behavior, achieves gains of approximately 0.1-0.3 under suboptimal behavior, and yields improvements of up to 0.4 under noisy optimal trajectories, while remaining competitive in fully optimal settings. This work contributes toward scalable and robust models for interpreting agent goals in uncertain environments.