Identifying the Training Stop Point with Noisy Labeled Data

Sree Ram Kamabattula, Venkat Devarajan, Babak Namazi, Ganesh Sankaranarayanan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Finding an early stopping point at maximum obtainable test accuracy (MOTA) is a challenging problem when training deep neural networks (DNNs) with noisy labeled data. Recent studies assume either that i) a clean validation set is available or ii) the noise ratio is known, or, both. However, often a clean validation set is unavailable, and the noise estimation can be inaccurate. To overcome these issues, we provide a novel training solution, free of these conditions. We analyze the rate of change of the training accuracy under different conditions to identify a training stop region. We further develop a heuristic algorithm (AutoTSP) to find a training stop point (TSP) at or close to MOTA. We validate the robustness of AutoTSP through several experiments on various datasets, noise ratios and architectures.

Original languageEnglish (US)
Title of host publicationProceedings - 2020 International Conference on Computational Science and Computational Intelligence, CSCI 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages457-463
Number of pages7
ISBN (Electronic)9781728176246
DOIs
StatePublished - Dec 2020
Externally publishedYes
Event2020 International Conference on Computational Science and Computational Intelligence, CSCI 2020 - Las Vegas, United States
Duration: Dec 16 2020Dec 18 2020

Publication series

NameProceedings - 2020 International Conference on Computational Science and Computational Intelligence, CSCI 2020

Conference

Conference2020 International Conference on Computational Science and Computational Intelligence, CSCI 2020
Country/TerritoryUnited States
CityLas Vegas
Period12/16/2012/18/20

Keywords

  • Memorization stages
  • MOTA
  • TSP

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Safety, Risk, Reliability and Quality

Fingerprint

Dive into the research topics of 'Identifying the Training Stop Point with Noisy Labeled Data'. Together they form a unique fingerprint.

Cite this