Learning from noisy labelsby importance reweighting: : a deep learning approach

Noisy labels could cause severe degradation to the classification performance. Especially for deep neural networks, noisy labels can be memorized and lead to poor generalization. Recently label noise robust deep learning has outperformed traditional shallow learning approaches in handling complex in...

Full description

Bibliographic Details
Main Author: Fang, Tongtong
Format: Others
Language:English
Published: KTH, Skolan för elektroteknik och datavetenskap (EECS) 2019
Subjects:
Online Access:http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-264125
id ndltd-UPSALLA1-oai-DiVA.org-kth-264125
record_format oai_dc
spelling ndltd-UPSALLA1-oai-DiVA.org-kth-2641252020-01-23T03:36:19ZLearning from noisy labelsby importance reweighting: : a deep learning approachengFang, TongtongKTH, Skolan för elektroteknik och datavetenskap (EECS)2019noisy labelimportance reweightingdeep learningdomain adaptationannoterad dataomviktningdjupt lärandedomänanpassningComputer and Information SciencesData- och informationsvetenskapNoisy labels could cause severe degradation to the classification performance. Especially for deep neural networks, noisy labels can be memorized and lead to poor generalization. Recently label noise robust deep learning has outperformed traditional shallow learning approaches in handling complex input data without prior knowledge of label noise generation. Learning from noisy labels by importance reweighting is well-studied. Existing work in this line using deep learning failed to provide reasonable importance reweighting criterion and thus got undesirable experimental performances. Targeting this knowledge gap and inspired by domain adaptation, we propose a novel label noise robust deep learning approach by importance reweighting. Noisy labeled training examples are weighted by minimizing the maximum mean discrepancy between the loss distributions of noisy labeled and clean labeled data. In experiments, the proposed approach outperforms other baselines. Results show a vast research potential of applying domain adaptation in label noise problem by bridging the two areas. Moreover, the proposed approach potentially motivate other interesting problems in domain adaptation by enabling importance reweighting to be used in deep learning. Felaktiga annoteringar kan sänka klassificeringsprestanda.Speciellt för djupa nätverk kan detta leda till dålig generalisering. Nyligen har brusrobust djup inlärning överträffat andra inlärningsmetoder när det gäller hantering av komplexa indata Befintligta resultat från djup inlärning kan dock inte tillhandahålla rimliga viktomfördelningskriterier. För att hantera detta kunskapsgap och inspirerat av domänanpassning föreslår vi en ny robust djup inlärningsmetod som använder omviktning. Omviktningen görs genom att minimera den maximala medelavvikelsen mellan förlustfördelningen av felmärkta och korrekt märkta data. I experiment slår den föreslagna metoden andra metoder. Resultaten visar en stor forskningspotential för att tillämpa domänanpassning. Dessutom motiverar den föreslagna metoden undersökningar av andra intressanta problem inom domänanpassning genom att möjliggöra smarta omviktningar. Student thesisinfo:eu-repo/semantics/bachelorThesistexthttp://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-264125TRITA-EECS-EX ; 2019:621application/pdfinfo:eu-repo/semantics/openAccess
collection NDLTD
language English
format Others
sources NDLTD
topic noisy label
importance reweighting
deep learning
domain adaptation
annoterad data
omviktning
djupt lärande
domänanpassning
Computer and Information Sciences
Data- och informationsvetenskap
spellingShingle noisy label
importance reweighting
deep learning
domain adaptation
annoterad data
omviktning
djupt lärande
domänanpassning
Computer and Information Sciences
Data- och informationsvetenskap
Fang, Tongtong
Learning from noisy labelsby importance reweighting: : a deep learning approach
description Noisy labels could cause severe degradation to the classification performance. Especially for deep neural networks, noisy labels can be memorized and lead to poor generalization. Recently label noise robust deep learning has outperformed traditional shallow learning approaches in handling complex input data without prior knowledge of label noise generation. Learning from noisy labels by importance reweighting is well-studied. Existing work in this line using deep learning failed to provide reasonable importance reweighting criterion and thus got undesirable experimental performances. Targeting this knowledge gap and inspired by domain adaptation, we propose a novel label noise robust deep learning approach by importance reweighting. Noisy labeled training examples are weighted by minimizing the maximum mean discrepancy between the loss distributions of noisy labeled and clean labeled data. In experiments, the proposed approach outperforms other baselines. Results show a vast research potential of applying domain adaptation in label noise problem by bridging the two areas. Moreover, the proposed approach potentially motivate other interesting problems in domain adaptation by enabling importance reweighting to be used in deep learning. === Felaktiga annoteringar kan sänka klassificeringsprestanda.Speciellt för djupa nätverk kan detta leda till dålig generalisering. Nyligen har brusrobust djup inlärning överträffat andra inlärningsmetoder när det gäller hantering av komplexa indata Befintligta resultat från djup inlärning kan dock inte tillhandahålla rimliga viktomfördelningskriterier. För att hantera detta kunskapsgap och inspirerat av domänanpassning föreslår vi en ny robust djup inlärningsmetod som använder omviktning. Omviktningen görs genom att minimera den maximala medelavvikelsen mellan förlustfördelningen av felmärkta och korrekt märkta data. I experiment slår den föreslagna metoden andra metoder. Resultaten visar en stor forskningspotential för att tillämpa domänanpassning. Dessutom motiverar den föreslagna metoden undersökningar av andra intressanta problem inom domänanpassning genom att möjliggöra smarta omviktningar.
author Fang, Tongtong
author_facet Fang, Tongtong
author_sort Fang, Tongtong
title Learning from noisy labelsby importance reweighting: : a deep learning approach
title_short Learning from noisy labelsby importance reweighting: : a deep learning approach
title_full Learning from noisy labelsby importance reweighting: : a deep learning approach
title_fullStr Learning from noisy labelsby importance reweighting: : a deep learning approach
title_full_unstemmed Learning from noisy labelsby importance reweighting: : a deep learning approach
title_sort learning from noisy labelsby importance reweighting: : a deep learning approach
publisher KTH, Skolan för elektroteknik och datavetenskap (EECS)
publishDate 2019
url http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-264125
work_keys_str_mv AT fangtongtong learningfromnoisylabelsbyimportancereweightingadeeplearningapproach
_version_ 1719309545621159936