BibLaTeX
@inproceedings{aristoteLearningWeightedAutomata2025,
author = {Aristote, Quentin and van Gool, Sam and Petrişan, Daniela
and Shirmohammadi, Mahsa},
publisher = {The Association for Computing Machinery},
title = {Learning {Weighted} {Automata} over {Number} {Rings,}
{Concretely} and {Categorically}},
booktitle = {2025 40th Annual ACM/IEEE Symposium on Logic in Computer
Science (LICS)},
pages = {417-430},
date = {2025-06},
urldate = {2025-04-28},
address = {New York, NY, USA},
url = {https://ieeexplore.ieee.org/document/11186332},
doi = {10.1109/LICS65433.2025.00038},
langid = {en},
abstract = {We develop a generic reduction procedure for active
learning problems. Our approach is inspired by a recent
polynomial-time reduction of the exact learning problem for weighted
automata over integers to that for weighted automata over rationals
(Buna-Marginean et al. 2024). Our procedure improves the efficiency
of a category-theoretic automata learning algorithm, and poses new
questions about the complexity of its implementation when
instantiated to concrete categories. As our second main
contribution, we address these complexity aspects in the concrete
setting of learning weighted automata over number rings, that is,
rings of integers in an algebraic number field. Assuming a full
representation of a number ring OK, we obtain an exact learning
algorithm of OK-weighted automata that runs in polynomial time in
the size of the target automaton, the logarithm of the length of the
longest counterexample, the degree of the number field, and the
logarithm of its discriminant. Our algorithm produces an automaton
that has at most one more state than the minimal one, and we prove
that doing better requires solving the principal ideal problem, for
which the best currently known algorithm is in quantum polynomial
time.}
}
BibTeX
@inproceedings{aristoteLearningWeightedAutomata2025,
author = {Aristote, Quentin and van Gool, Sam and Petrişan, Daniela
and Shirmohammadi, Mahsa},
publisher = {The Association for Computing Machinery},
title = {Learning {Weighted} {Automata} over {Number} {Rings,}
{Concretely} and {Categorically}},
booktitle = {2025 40th Annual ACM/IEEE Symposium on Logic in Computer
Science (LICS)},
pages = {417-430},
year = {2025},
month = {jun},
address = {New York, NY, USA},
url = {https://ieeexplore.ieee.org/document/11186332}
}
CSL JSON
{"id":"aristoteLearningWeightedAutomata2025","abstract":"We develop a generic reduction procedure for active learning problems. Our approach is inspired by a recent polynomial-time reduction of the exact learning problem for weighted automata over integers to that for weighted automata over rationals (Buna-Marginean et al. 2024). Our procedure improves the efficiency of a category-theoretic automata learning algorithm, and poses new questions about the complexity of its implementation when instantiated to concrete categories. \n\nAs our second main contribution, we address these complexity aspects in the concrete setting of learning weighted automata over number rings, that is, rings of integers in an algebraic number field. Assuming a full representation of a number ring OK, we obtain an exact learning algorithm of OK-weighted automata that runs in polynomial time in the size of the target automaton, the logarithm of the length of the longest counterexample, the degree of the number field, and the logarithm of its discriminant. Our algorithm produces an automaton that has at most one more state than the minimal one, and we prove that doing better requires solving the principal ideal problem, for which the best currently known algorithm is in quantum polynomial time.","accessed":{"date-parts":[["2025",4,28]]},"author":[{"family":"Aristote","given":"Quentin"},{"family":"Gool","given":"Sam","non-dropping-particle":"van"},{"family":"Petrişan","given":"Daniela"},{"family":"Shirmohammadi","given":"Mahsa"}],"citation-key":"aristoteLearningWeightedAutomata2025","container-title":"2025 40th Annual ACM/IEEE Symposium on Logic in Computer Science (LICS)","DOI":"10.1109/LICS65433.2025.00038","event-place":"New York, NY, USA","event-title":"Logics in Computer Science (LICS)","issued":{"date-parts":[["2025",6]]},"language":"en","page":"417-430","publisher":"The Association for Computing Machinery","publisher-place":"New York, NY, USA","title":"Learning Weighted Automata over Number Rings, Concretely and Categorically","type":"paper-conference","URL":"https://ieeexplore.ieee.org/document/11186332"}