# A Novel Self-Supervised Re-labeling Approach for Training with Noisy Labels

arXiv: https://arxiv.org/abs/1910.05700 · HTML: https://mlsanity.com/paper/1910.05700/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Oct 2019.

## Built on by (most cited first, 10 of 10)

- [Co-learning: Learning from Noisy Labels with Self-supervision](https://mlsanity.com/paper/2108.04063/) (2021)
- [Joint-Modal Label Denoising for Weakly-Supervised Audio-Visual Video Parsing](https://mlsanity.com/paper/2204.11573/) (2022)
- [Continual Learning on Noisy Data Streams via Self-Purified Replay](https://mlsanity.com/paper/2110.07735/) (2021)
- [Estimating the electrical power output of industrial devices with end-to-end time-series classification in the presence of label noise](https://mlsanity.com/paper/2105.00349/) (2021)
- [Foster Adaptivity and Balance in Learning with Noisy Labels](https://mlsanity.com/paper/2407.02778/) (2024)
- [Learning with Imbalanced Noisy Data by Preventing Bias in Sample Selection](https://mlsanity.com/paper/2402.11242/) (2024)
- [Do We Really Need Gold Samples for Sample Weighting Under Label Noise?](https://mlsanity.com/paper/2104.09045/) (2021)
- [Entailment as Robust Self-Learner](https://mlsanity.com/paper/2305.17197/) (2023)
- [Friends and Foes in Learning from Noisy Labels](https://mlsanity.com/paper/2103.15055/) (2021)
- [Leveraging Data Symmetries to Select an Optimal Subset of Training Data under Label Noise](https://mlsanity.com/paper/2605.01874/) (2026)

## Abstract (arXiv)

The major driving force behind the immense success of deep learning models is the availability of large datasets along with their clean labels. Unfortunately, this is very difficult to obtain, which has motivated research on the training of deep models in the presence of label noise and ways to avoid over-fitting on the noisy labels. In this work, we build upon the seminal work in this area, Co-teaching and propose a simple, yet efficient approach termed mCT-S2R (modified co-teaching with self-supervision and re-labeling) for this task. First, to deal with significant amount of noise in the labels, we propose to use self-supervision to generate robust features without using any labels. Next, using a parallel network architecture, an estimate of the clean labeled portion of the data is obtained. Finally, using this data, a portion of the estimated noisy labeled portion is re-labeled, before resuming the network training with the augmented data. Extensive experiments on three standard datasets show the effectiveness of the proposed framework.