@INPROCEEDINGS{Amado2021,
  author = {Leonardo R. Amado and Ramon F. Pereira and Felipe Meneguzzi},
  title = {Combining LSTMs and Symbolic Approaches for Robust Plan Recognition},
  booktitle = {Proceedings of the Twentieth International Conference on Autonomous
	Agents and Multiagent Systems},
  year = {2021},
  pages = { },
abstract = {Plan recognition is the task of identifying the complete plan an agent is performing to reach a certain goal by observing its interactions in an environment. 
Regardless of the underlying techniques, most plan recognition approaches are directly affected by the quality of the given observations. 
To mitigate such issue, we combine learning techniques and landmark extraction algorithms into an approach for plan recognition that compensates for noise and missing observations using prior data. 
We evaluate our approach in standard human-designed planning domains as well as domain theories automatically generated from real-world data. 
Our evaluation shows that our approach reliably computes correct plans in all domains of the experimental dataset, in both low observability problems and problems with noisy observation, outperforming approaches that rely exclusively on the domain theory.}
}