Generalization bounds for learning weighted automata

Balle, Borja and Mohri, Mehryar (2018) Generalization bounds for learning weighted automata. Theoretical Computer Science, 716. pp. 89-106. ISSN 0304-3975

Full text not available from this repository.

Abstract

This paper studies the problem of learning weighted automata from a finite sample of strings with real-valued labels. We consider several hypothesis classes of weighted automata defined in terms of three different measures: the norm of an automaton's weights, the norm of the function computed by an automaton, and the norm of the corresponding Hankel matrix. We present new data-dependent generalization guarantees for learning weighted automata expressed in terms of the Rademacher complexity of these classes. We further present upper bounds on these Rademacher complexities, which reveal key new data-dependent terms related to the complexity of learning weighted automata.

Item Type: Journal Article
Journal or Publication Title: Theoretical Computer Science
Uncontrolled Keywords: /dk/atira/pure/subjectarea/asjc/1700
Subjects:
Departments: Faculty of Science and Technology > Mathematics and Statistics
ID Code: 89047
Deposited By: ep_importer_pure
Deposited On: 19 Dec 2018 09:40
Refereed?: Yes
Published?: Published
Last Modified: 11 Feb 2020 03:48
URI: https://eprints.lancs.ac.uk/id/eprint/89047

Actions (login required)

View Item View Item