Training Artificial Neural Networks by Generalized Likelihood Ratio Method: Exploring Brain-like Learning to Improve Robustness

Training Artificial Neural Networks by Generalized Likelihood Ratio Method: Exploring Brain-like Learning to Improve Robustness

In this work, we propose a generalized likelihood ratio method capable of training the artificial neural networks with some biological brain-like mechanisms,.e.g., (a) learning by the loss value, (b) learning via neurons with discontinuous activation and loss functions. The traditional back propagat...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Xiao, Li, Peng, Yijie, Hong, Jeff, Ke, Zewu, Yang, Shuhuai
Format: Artikel
Sprache:eng
Schlagworte:
Computer Science - Artificial Intelligence
Computer Science - Learning
Statistics - Machine Learning
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Xiao, Li
Peng, Yijie
Hong, Jeff
Ke, Zewu
Yang, Shuhuai
description In this work, we propose a generalized likelihood ratio method capable of training the artificial neural networks with some biological brain-like mechanisms,.e.g., (a) learning by the loss value, (b) learning via neurons with discontinuous activation and loss functions. The traditional back propagation method cannot train the artificial neural networks with aforementioned brain-like learning mechanisms. Numerical results show that the robustness of various artificial neural networks trained by the new method is significantly improved when the input data is affected by both the natural noises and adversarial attacks. Code is available: \url{https://github.com/LX-doctorAI/GLR_ADV} .
doi_str_mv 10.48550/arxiv.1902.00358
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1902_00358</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1902_00358</sourcerecordid><originalsourceid>FETCH-LOGICAL-a678-b9ff8ed2b34a43fab8aacb409dfd47c740c3386ac4f1b674ac538f4d834bdacf3</originalsourceid><addsrcrecordid>eNotkE1uwjAYRL3poqI9QFf4Akkd7CROdxRRipS2Eso--vwHFiGOHEOhm14dknY10mj0RnoIPSUkZjxNyTP4sz3FSUFmMSE05ffot_JgW9tu8dwHa6y00OBPffRjhG_n9z0WF7zSrb519kcrXNq9buzOOYU3EKzDHzrsnHrBy3PXOD_AXgdq1NyGuNTgx4Pg8PrQeXfSeOPEsQ-t7vsHdGeg6fXjf05Q9basFu9R-bVaL-ZlBFnOI1EYw7WaCcqAUQOCA0jBSKGMYrnMGZGU8gwkM4nIcgYypdwwxSkTCqShEzT9w44G6s7bA_hLPZioRxP0CoqUXL8</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Training Artificial Neural Networks by Generalized Likelihood Ratio Method: Exploring Brain-like Learning to Improve Robustness</title><source>arXiv.org</source><creator>Xiao, Li ; Peng, Yijie ; Hong, Jeff ; Ke, Zewu ; Yang, Shuhuai</creator><creatorcontrib>Xiao, Li ; Peng, Yijie ; Hong, Jeff ; Ke, Zewu ; Yang, Shuhuai</creatorcontrib><description>In this work, we propose a generalized likelihood ratio method capable of training the artificial neural networks with some biological brain-like mechanisms,.e.g., (a) learning by the loss value, (b) learning via neurons with discontinuous activation and loss functions. The traditional back propagation method cannot train the artificial neural networks with aforementioned brain-like learning mechanisms. Numerical results show that the robustness of various artificial neural networks trained by the new method is significantly improved when the input data is affected by both the natural noises and adversarial attacks. Code is available: \url{https://github.com/LX-doctorAI/GLR_ADV} .</description><identifier>DOI: 10.48550/arxiv.1902.00358</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Learning ; Statistics - Machine Learning</subject><creationdate>2019-01</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/1902.00358$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1902.00358$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Xiao, Li</creatorcontrib><creatorcontrib>Peng, Yijie</creatorcontrib><creatorcontrib>Hong, Jeff</creatorcontrib><creatorcontrib>Ke, Zewu</creatorcontrib><creatorcontrib>Yang, Shuhuai</creatorcontrib><title>Training Artificial Neural Networks by Generalized Likelihood Ratio Method: Exploring Brain-like Learning to Improve Robustness</title><description>In this work, we propose a generalized likelihood ratio method capable of training the artificial neural networks with some biological brain-like mechanisms,.e.g., (a) learning by the loss value, (b) learning via neurons with discontinuous activation and loss functions. The traditional back propagation method cannot train the artificial neural networks with aforementioned brain-like learning mechanisms. Numerical results show that the robustness of various artificial neural networks trained by the new method is significantly improved when the input data is affected by both the natural noises and adversarial attacks. Code is available: \url{https://github.com/LX-doctorAI/GLR_ADV} .</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Learning</subject><subject>Statistics - Machine Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotkE1uwjAYRL3poqI9QFf4Akkd7CROdxRRipS2Eso--vwHFiGOHEOhm14dknY10mj0RnoIPSUkZjxNyTP4sz3FSUFmMSE05ffot_JgW9tu8dwHa6y00OBPffRjhG_n9z0WF7zSrb519kcrXNq9buzOOYU3EKzDHzrsnHrBy3PXOD_AXgdq1NyGuNTgx4Pg8PrQeXfSeOPEsQ-t7vsHdGeg6fXjf05Q9basFu9R-bVaL-ZlBFnOI1EYw7WaCcqAUQOCA0jBSKGMYrnMGZGU8gwkM4nIcgYypdwwxSkTCqShEzT9w44G6s7bA_hLPZioRxP0CoqUXL8</recordid><startdate>20190131</startdate><enddate>20190131</enddate><creator>Xiao, Li</creator><creator>Peng, Yijie</creator><creator>Hong, Jeff</creator><creator>Ke, Zewu</creator><creator>Yang, Shuhuai</creator><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20190131</creationdate><title>Training Artificial Neural Networks by Generalized Likelihood Ratio Method: Exploring Brain-like Learning to Improve Robustness</title><author>Xiao, Li ; Peng, Yijie ; Hong, Jeff ; Ke, Zewu ; Yang, Shuhuai</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a678-b9ff8ed2b34a43fab8aacb409dfd47c740c3386ac4f1b674ac538f4d834bdacf3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Learning</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Xiao, Li</creatorcontrib><creatorcontrib>Peng, Yijie</creatorcontrib><creatorcontrib>Hong, Jeff</creatorcontrib><creatorcontrib>Ke, Zewu</creatorcontrib><creatorcontrib>Yang, Shuhuai</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Xiao, Li</au><au>Peng, Yijie</au><au>Hong, Jeff</au><au>Ke, Zewu</au><au>Yang, Shuhuai</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Training Artificial Neural Networks by Generalized Likelihood Ratio Method: Exploring Brain-like Learning to Improve Robustness</atitle><date>2019-01-31</date><risdate>2019</risdate><abstract>In this work, we propose a generalized likelihood ratio method capable of training the artificial neural networks with some biological brain-like mechanisms,.e.g., (a) learning by the loss value, (b) learning via neurons with discontinuous activation and loss functions. The traditional back propagation method cannot train the artificial neural networks with aforementioned brain-like learning mechanisms. Numerical results show that the robustness of various artificial neural networks trained by the new method is significantly improved when the input data is affected by both the natural noises and adversarial attacks. Code is available: \url{https://github.com/LX-doctorAI/GLR_ADV} .</abstract><doi>10.48550/arxiv.1902.00358</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.1902.00358
ispartof
issn
language eng
recordid cdi_arxiv_primary_1902_00358
source arXiv.org
subjects Computer Science - Artificial Intelligence
Computer Science - Learning
Statistics - Machine Learning
title Training Artificial Neural Networks by Generalized Likelihood Ratio Method: Exploring Brain-like Learning to Improve Robustness
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-01T03%3A41%3A03IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Training%20Artificial%20Neural%20Networks%20by%20Generalized%20Likelihood%20Ratio%20Method:%20Exploring%20Brain-like%20Learning%20to%20Improve%20Robustness&rft.au=Xiao,%20Li&rft.date=2019-01-31&rft_id=info:doi/10.48550/arxiv.1902.00358&rft_dat=%3Carxiv_GOX%3E1902_00358%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true