Learning DFA from Correction and Equivalence Queries

2006 ◽  
pp. 281-292 ◽  
Author(s):  
Leonor Becerra-Bonache ◽  
Adrian Horia Dediu ◽  
Cristina Tîrnăucă
Keyword(s):  
Equivalence Queries

ORACLES IN $\Sigma^p_2$ ARE SUFFFICIENT FOR EXACT LEARNING

2000 ◽  
Vol 11 (04) ◽  
pp. 613-632 ◽  
Author(s):  
Johannes Köbler ◽  
Wolfgang Lindner
Keyword(s):  
Polynomial Time ◽  
Learning Model ◽  
Query Complexity ◽  
Boolean Circuits ◽  
Membership Queries ◽  
Exact Learning ◽  
Equivalence Queries

We study the learnability of representation classes in Angluin's exact learning model. In particular, we consider the following three query types: equivalence queries, equivalence and membership queries, and membership queries only. We show in all three cases that polynomial query complexity implies already polynomial-time learnability, provided that the learner additionally has access to an oracle in [Formula: see text]. It follows that boolean circuits are polynomial-time learnable with equivalence queries and the help of an oracle in [Formula: see text].a

scholarly journals Weighted Automata Extraction from Recurrent Neural Networks via Regression on State Spaces

2020 ◽  
Vol 34 (04) ◽  
pp. 5306-5314
Author(s):  
Takamasa Okudono ◽  
Masaki Waga ◽  
Taro Sekiyama ◽  
Ichiro Hasuo
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Recurrent Neural Network ◽  
Recurrent Neural Networks ◽  
Learning Algorithm ◽  
Internal State ◽  
State Spaces ◽  
Regression Methods ◽  
Weighted Automata ◽  
Equivalence Queries

We present a method to extract a weighted finite automaton (WFA) from a recurrent neural network (RNN). Our method is based on the WFA learning algorithm by Balle and Mohri, which is in turn an extension of Angluin's classic L* algorithm. Our technical novelty is in the use of regression methods for the so-called equivalence queries, thus exploiting the internal state space of an RNN to prioritize counterexample candidates. This way we achieve a quantitative/weighted extension of the recent work by Weiss, Goldberg and Yahav that extracts DFAs. We experimentally evaluate the accuracy, expressivity and efficiency of the extracted WFAs.

scholarly journals Actively Learning Concepts and Conjunctive Queries under ELr-Ontologies

2021 ◽  
Author(s):  
Maurice Funk ◽  
Jean Christoph Jung ◽  
Carsten Lutz
Keyword(s):  
Active Learning ◽  
Polynomial Time ◽  
Description Logic ◽  
Learning Algorithm ◽  
Domain Expert ◽  
Conjunctive Queries ◽  
Membership Queries ◽  
Equivalence Queries

We consider the problem to learn a concept or a query in the presence of an ontology formulated in the description logic ELr, in Angluin's framework of active learning that allows the learning algorithm to interactively query an oracle (such as a domain expert). We show that the following can be learned in polynomial time: (1) EL-concepts, (2) symmetry-free ELI-concepts, and (3) conjunctive queries (CQs) that are chordal, symmetry-free, and of bounded arity. In all cases, the learner can pose to the oracle membership queries based on ABoxes and equivalence queries that ask whether a given concept/query from the considered class is equivalent to the target. The restriction to bounded arity in (3) can be removed when we admit unrestricted CQs in equivalence queries. We also show that EL-concepts are not polynomial query learnable in the presence of ELI-ontologies.

Sign in / Sign up

Export Citation Format

Share Document