Robust Self-Training with Closed-loop Label Correction for Learning from Noisy Labels
About
Training deep neural networks with noisy labels remains a significant challenge, often leading to degraded performance. Existing methods for handling label noise typically rely on either transition matrix, noise detection, or meta-learning techniques, but they often exhibit low utilization efficiency of noisy samples and incur high computational costs. In this paper, we propose a self-training label correction framework using decoupled bilevel optimization, where a classifier and neural correction function co-evolve. Leveraging a small clean dataset, our method employs noisy posterior simulation and intermediate features to transfer ground-truth knowledge, forming a closed-loop feedback system that prevents error amplification. Theoretical guarantees underpin the stability of our approach, and extensive experiments on benchmark datasets like CIFAR and Clothing1M confirm state-of-the-art performance with reduced training time, highlighting its practical applicability for learning from noisy labels.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Image Classification | Clothing1M (test) | Accuracy80.23 | 574 | |
| Image Classification | CIFAR-10 (test) | Test Accuracy (Symmetric, η=0.2)92.48 | 12 | |
| Image Classification | CIFAR-100 (test) | Test Accuracy (Symmetric, η=0.2)68.73 | 12 | |
| Image Classification | CIFAR-100 Instance-dependent Noise (1k clean samples) | Accuracy (eta=0.2)67.42 | 10 | |
| Image Classification | CIFAR-100 with Symmetric Noise (1k clean samples) | Accuracy (eta=0.2)68.13 | 10 | |
| Image Classification | CIFAR-100 with Asymmetric Noise (1k clean samples) | Accuracy (eta=0.2)69.92 | 10 |