Please use this identifier to cite or link to this item: http://hdl.handle.net/10071/26524
Author(s): Bacellar, A. T. L.
Susskind, Z.
Villon, L. A. Q.
Miranda, I. D. S.
Araújo, L. S. de.
Dutra, D. L. C.
Breternitz Jr, M.
John, L. K.
Lima, P. M. V.
França, F. M. G.
Date: 2022
Title: Distributive thermometer: A new unary encoding for weightless neural networks
Book title/volume: ESANN 2022 proceedings
Pages: 31 - 36
Event title: 30th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
Reference: Bacellar, A. T. L., Susskind, Z., Villon, L. A. Q., Miranda, I. D. S., Araújo, L. S. de., Dutra, D. L. C., Breternitz Jr., M., John, L. K., Lima, P. M. V., & França, F. M. G. (2022). Distributive thermometer: A new unary encoding for weightless neural networks. In ESANN 2022 proceedings (pp. 31-36). https://doi.org/10.14428/esann/2022.ES2022-94
ISBN: 978287587 084-1
DOI (Digital Object Identifier): 10.14428/esann/2022.ES2022-94
Abstract: The binary encoding of real valued inputs is a crucial part of Weightless Neural Networks. The Linear Thermometer and its variations are the most prominent methods to determine binary encoding for input data but, as they make assumptions about the input distribution, the resulting encoding is sub-optimal and possibly wasteful when the assumption is incorrect. We propose a new thermometer approach that doesn’t require such assumptions. Our results show that it achieves similar or better accuracy when compared to a thermometer that correctly assumes the distribution, and accuracy gains up to 26.3% when other thermometer representations assume an unsound distribution.
Peerreviewed: yes
Access type: Open Access
Appears in Collections:ISTAR-CRI - Comunicações a conferências internacionais

Files in This Item:
File SizeFormat 
conferenceobject_91314.pdf1,54 MBAdobe PDFView/Open


FacebookTwitterDeliciousLinkedInDiggGoogle BookmarksMySpaceOrkut
Formato BibTex mendeley Endnote Logotipo do DeGóis Logotipo do Orcid 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.