Towards Efficient Active Learning of PDFA
Formal Languages and Automata Theory
2022-06-22 v1 Artificial Intelligence
Machine Learning
Abstract
We propose a new active learning algorithm for PDFA based on three main aspects: a congruence over states which takes into account next-symbol probability distributions, a quantization that copes with differences in distributions, and an efficient tree-based data structure. Experiments showed significant performance gains with respect to reference implementations.
Cite
@article{arxiv.2206.09004,
title = {Towards Efficient Active Learning of PDFA},
author = {Franz Mayr and Sergio Yovine and Federico Pan and Nicolas Basset and Thao Dang},
journal= {arXiv preprint arXiv:2206.09004},
year = {2022}
}
Comments
11 pages, 7 figures, workshop paper