A Framework using Contrastive Learning for Classification with Noisy Labels
This work addresses noisy label problems in image classification, but it is incremental as it combines existing strategies into a framework.
The paper tackles image classification with noisy labels by using contrastive learning as a pre-training step, showing that it significantly boosts robustness across various loss functions, with experiments on benchmarks and real-world datasets confirming performance gains.
We propose a framework using contrastive learning as a pre-training task to perform image classification in the presence of noisy labels. Recent strategies such as pseudo-labeling, sample selection with Gaussian Mixture models, weighted supervised contrastive learning have been combined into a fine-tuning phase following the pre-training. This paper provides an extensive empirical study showing that a preliminary contrastive learning step brings a significant gain in performance when using different loss functions: non-robust, robust, and early-learning regularized. Our experiments performed on standard benchmarks and real-world datasets demonstrate that: i) the contrastive pre-training increases the robustness of any loss function to noisy labels and ii) the additional fine-tuning phase can further improve accuracy but at the cost of additional complexity.