Learning from Noisy Labels with Decoupled Meta Label Purifier

TL;DR

This paper introduces DMLP, a multi-stage label purifier that decouples feature learning from label correction, significantly improving deep neural network training with noisy labels by focusing on discriminative features and effective label correction.

Contribution

The paper proposes a novel decoupled multi-stage label purification method, DMLP, which enhances noisy label learning by separating feature extraction from label correction.

Findings

DMLP achieves state-of-the-art results on synthetic noisy datasets.

DMLP performs well under high noise levels.

Decoupling improves label correction accuracy.

Abstract

Training deep neural networks(DNN) with noisy labels is challenging since DNN can easily memorize inaccurate labels, leading to poor generalization ability. Recently, the meta-learning based label correction strategy is widely adopted to tackle this problem via identifying and correcting potential noisy labels with the help of a small set of clean validation data. Although training with purified labels can effectively improve performance, solving the meta-learning problem inevitably involves a nested loop of bi-level optimization between model weights and hyper-parameters (i.e., label distribution). As compromise, previous methods resort to a coupled learning process with alternating update. In this paper, we empirically find such simultaneous optimization over both model weights and label distribution can not achieve an optimal routine, consequently limiting the representation ability of backbone and accuracy of corrected labels. From this observation, a novel multi-stage label purifier named DMLP is proposed. DMLP decouples the label correction process into label-free representation learning and a simple meta label purifier. In this way, DMLP can focus on extracting discriminative feature and label correction in two distinctive stages. DMLP is a plug-and-play label purifier, the purified labels can be directly reused in naive end-to-end network retraining or other robust learning methods, where state-of-the-art results are obtained on several synthetic and real-world noisy datasets, especially under high noise levels.