Author
Thakur, A
Abrol, V
Sharma, P
Rajan, P
Assoc, I
Journal title
19TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2018), VOLS 1-6
DOI
10.21437/Interspeech.2018-1705
Volume
2018-September
Last updated
2021-02-15T00:07:21.427+00:00
Page
2127-2131
Abstract
© 2018 International Speech Communication Association. All rights reserved. In this paper, a deep convex matrix factorization framework is proposed for bioacoustics classification. Archetypal analysis, a form of convex non-negative matrix factorization, is used for acoustic modelling at each level of this framework. At first level, the input feature matrix is factorized into an archetypal dictionary and corresponding convex representations. The representation matrix obtained at the first level is further factorized into a dictionary and convex representations at second level. This hierarchical factorization continues until a desired depth is achieved. We observe that the dictionaries at different levels model complimentary information present in the data. The atoms of the dictionary learned at the first layer lie on convex hull of the data, thus try to model the extremal behaviour. On the contrary, atoms of the deeper dictionaries lie on the convex hull as well as inside the convex hull. Hence, these dictionaries can simultaneously model the extremal and average behaviour of the data. The convex representations obtained from these deeper dictionaries are referred as deep convex representations. Due to inherent sparsity, they result in efficient classification performance. Through experiments on two available bioacoustics datasets, we show that the proposed approach yield comparable or better results than state-of-art approaches.
Symplectic ID
925641
Download URL
http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=WOS:000465363900446&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=4fd6f7d59a501f9b8bac2be37914c43e
Publication type
Conference Paper
ISBN-13
978-1-5108-7221-9
Publication date
2018
Please contact us with feedback and comments about this page. Created on 04 Feb 2019 - 14:11.