Discriminatively Reranking Abductive Proofs for Plan Recognition


  • Sam Wiseman Harvard University
  • Stuart Shieber Harvard University




We investigate the use of a simple, discriminative reranking approach to plan recognition in an abductive setting. In contrast to recent work, which attempts to model abductive plan recognition using various formalisms that integrate logic and graphical models (such as Markov Logic Networks or Bayesian Logic Programs), we instead advocate a simpler, more flexible approach in which plans found through an abductive beam-search are discriminatively scored based on arbitrary features. We show that this approach performs well even with relatively few positive training examples, and we obtain state-of-the-art results on two abductive plan recognition datasets, outperforming more complicated systems.




How to Cite

Wiseman, S., & Shieber, S. (2014). Discriminatively Reranking Abductive Proofs for Plan Recognition. Proceedings of the International Conference on Automated Planning and Scheduling, 24(1), 380-384. https://doi.org/10.1609/icaps.v24i1.13679