GRAML: Goal Recognition As Metric Learning
This addresses the need for automated and adaptable goal recognition in AI systems, offering a solution that reduces training time for new goals, though it is incremental in improving existing data-driven approaches.
The paper tackles the problem of goal recognition (GR) from observed actions by introducing GRAML, which treats GR as a metric learning task using a Siamese network and RNN to learn embeddings that distinguish between goals, enabling quick adaptation to new goals with minimal examples. It demonstrates speed, flexibility, and runtime improvements over state-of-the-art methods while maintaining accurate recognition.
Goal Recognition (GR) is the problem of recognizing an agent's objectives based on observed actions. Recent data-driven approaches for GR alleviate the need for costly, manually crafted domain models. However, these approaches can only reason about a pre-defined set of goals, and time-consuming training is needed for new emerging goals. To keep this model-learning automated while enabling quick adaptation to new goals, this paper introduces GRAML: Goal Recognition As Metric Learning. GRAML uses a Siamese network to treat GR as a deep metric learning task, employing an RNN that learns a metric over an embedding space, where the embeddings for observation traces leading to different goals are distant, and embeddings of traces leading to the same goals are close. This metric is especially useful when adapting to new goals, even if given just one example observation trace per goal. Evaluated on a versatile set of environments, GRAML shows speed, flexibility, and runtime improvements over the state-of-the-art GR while maintaining accurate recognition.