Goal Recognition via Linear Programming

Felipe Meneguzzi,Luísa R. de A. Santos,Ramon Fraga Pereira,André G. Pereira

2024-04-12

Abstract:Goal Recognition is the task by which an observer aims to discern the goals that correspond to plans that comply with the perceived behavior of subject agents given as a sequence of observations. Research on Goal Recognition as Planning encompasses reasoning about the model of a planning task, the observations, and the goals using planning techniques, resulting in very efficient recognition approaches. In this article, we design novel recognition approaches that rely on the Operator-Counting framework, proposing new constraints, and analyze their constraints' properties both theoretically and empirically. The Operator-Counting framework is a technique that efficiently computes heuristic estimates of cost-to-goal using Integer/Linear Programming (IP/LP). In the realm of theory, we prove that the new constraints provide lower bounds on the cost of plans that comply with observations. We also provide an extensive empirical evaluation to assess how the new constraints improve the quality of the solution, and we found that they are especially informed in deciding which goals are unlikely to be part of the solution. Our novel recognition approaches have two pivotal advantages: first, they employ new IP/LP constraints for efficiently recognizing goals; second, we show how the new IP/LP constraints can improve the recognition of goals under both partial and noisy observability.

Artificial Intelligence,Human-Computer Interaction

What problem does this paper attempt to address?

The paper attempts to address the problem of improving recognition accuracy and efficiency in target recognition tasks through linear programming methods. Specifically, the authors design a new method based on the Operator-Counting framework, proposing new constraints in the form of Integer Programming or Linear Programming (IP/LP), and analyze the theoretical and empirical properties of these constraints. The main contributions include:

1. **Theoretical Proof**: Proving that the newly proposed constraints can provide a lower bound on the plan cost that conforms to observed behavior.
2. **Improved Recognition Quality**: Demonstrating through extensive experiments that the new constraints can significantly enhance the quality of solutions, especially under partially observable and noisy observation conditions.
3. **Application of New Constraints**: Proposing new constraints based on linear programming/integer programming (such as landmark constraints) and utilizing these constraints to extract information in target recognition tasks.

In summary, the paper aims to improve the shortcomings of existing techniques by introducing and studying new target recognition methods based on linear programming, particularly in handling incomplete and noisy data.