Fact Probability Vector Based Goal Recognition

TL;DR

This paper introduces a novel goal recognition method using fact probability vectors that compares observed facts with expected probabilities, providing improved accuracy and efficiency over existing techniques.

Contribution

It proposes a new vector-based approach for goal recognition that approximates expected fact probabilities and demonstrates empirical improvements over state-of-the-art methods.

Findings

Enhanced goal recognition accuracy

Reduced computational complexity

Effective approximation of fact probabilities

Abstract

We present a new approach to goal recognition that involves comparing observed facts with their expected probabilities. These probabilities depend on a specified goal g and initial state s0. Our method maps these probabilities and observed facts into a real vector space to compute heuristic values for potential goals. These values estimate the likelihood of a given goal being the true objective of the observed agent. As obtaining exact expected probabilities for observed facts in an observation sequence is often practically infeasible, we propose and empirically validate a method for approximating these probabilities. Our empirical results show that the proposed approach offers improved goal recognition precision compared to state-of-the-art techniques while reducing computational complexity.