Number of the records: 1
Classification by Sparse Neural Networks
- 1.
SYSNO ASEP 0485611 Document Type J - Journal Article R&D Document Type Journal Article Subsidiary J Článek ve WOS Title Classification by Sparse Neural Networks Author(s) Kůrková, Věra (UIVT-O) RID, SAI, ORCID
Sanguineti, M. (IT)Source Title IEEE Transactions on Neural Networks and Learning Systems - ISSN 2162-237X
Roč. 30, č. 9 (2019), s. 2746-2754Number of pages 9 s. Language eng - English Country US - United States Keywords Binary classification ; Chernoff–Hoeffding bound ; dictionaries of computational units ; feedforward networks ; measures of sparsity Subject RIV IN - Informatics, Computer Science OECD category Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8) R&D Projects GA15-18108S GA ČR - Czech Science Foundation (CSF) GA18-23827S GA ČR - Czech Science Foundation (CSF) Method of publishing Limited access Institutional support UIVT-O - RVO:67985807 UT WOS 000482589400015 EID SCOPUS 85071708566 DOI 10.1109/TNNLS.2018.2888517 Annotation The choice of dictionaries of computational units suitable for efficient computation of binary classification tasks is investigated. To deal with exponentially growing sets of tasks with increasingly large domains, a probabilistic model is introduced. The relevance of tasks for a given application area is modeled by a product probability distribution on the set of all binary-valued functions. Approximate measures of network sparsity are studied in terms of variational norms tailored to dictionaries of computational units. Bounds on these norms are proven using the Chernoff–Hoeffding bound on sums of independent random variables that need not be identically distributed. Consequences of the probabilistic results for the choice of dictionaries of computational units are derived. It is shown that when a priori knowledge of a type of classification tasks is limited, then the sparsity may be achieved only at the expense of large sizes of dictionaries. Workplace Institute of Computer Science Contact Tereza Šírová, sirova@cs.cas.cz, Tel.: 266 053 800 Year of Publishing 2020 Electronic address http://dx.doi.org/10.1109/TNNLS.2018.2888517
Number of the records: 1