Learning with Noisy Labels by Efficient Transition Matrix Estimation to Combat Label Miscorrection

29 Nov 2021  ·  Seong Min Kye, Kwanghee Choi, Joonyoung Yi, Buru Chang ·

Recent studies on learning with noisy labels have shown remarkable performance by exploiting a small clean dataset. In particular, model agnostic meta-learning-based label correction methods further improve performance by correcting noisy labels on the fly. However, there is no safeguard on the label miscorrection, resulting in unavoidable performance degradation. Moreover, every training step requires at least three back-propagations, significantly slowing down the training speed. To mitigate these issues, we propose a robust and efficient method that learns a label transition matrix on the fly. Employing the transition matrix makes the classifier skeptical about all the corrected samples, which alleviates the miscorrection issue. We also introduce a two-head architecture to efficiently estimate the label transition matrix every iteration within a single back-propagation, so that the estimated matrix closely follows the shifting noise distribution induced by label correction. Extensive experiments demonstrate that our approach shows the best performance in training efficiency while having comparable or better accuracy than existing methods.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Classification Clothing1M (using clean data) FasTEN Accuracy 77.83% # 6

Methods


No methods listed for this paper. Add relevant methods here