Improving Label Noise Robustness with Data Augmentation and Semi-Supervised Learning (Student Abstract)

Authors

  • Kento Nishi Lynbrook High School, San Jose, CA 95129
  • Yi Ding University of California, Santa Barbara
  • Alex Rich University of California, Santa Barbara
  • Tobias Höllerer University of California, Santa Barbara

Keywords:

Learning With Noisy Labels, Label Noise Robustness, Data Augmentation, Semi-Supervised Learning, Image Classification

Abstract

Modern machine learning algorithms typically require large amounts of labeled training data to fit a reliable model. To minimize the cost of data collection, researchers often employ techniques such as crowdsourcing and web scraping. However, web data and human annotations are known to exhibit high margins of error, resulting in sizable amounts of incorrect labels. Poorly labeled training data can cause models to overfit to the noise distribution, crippling performance in real-world applications. In this work, we investigate the viability of using data augmentation in conjunction with semi-supervised learning to improve the label noise robustness of image classification models. We conduct several experiments using noisy variants of the CIFAR-10 image classification dataset to benchmark our method against existing algorithms. Experimental results show that our augmentative SSL approach improves upon the state-of-the-art.

Downloads

Published

2021-05-18

How to Cite

Nishi, K., Ding, Y., Rich, A., & Höllerer, T. (2021). Improving Label Noise Robustness with Data Augmentation and Semi-Supervised Learning (Student Abstract). Proceedings of the AAAI Conference on Artificial Intelligence, 35(18), 15855-15856. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/17924

Issue

Section

AAAI Student Abstract and Poster Program