Potential Energy based Mixture Model for Noisy Label Learning
This addresses noisy label learning for deep neural networks, which is an incremental improvement focusing on incorporating data structure.
The paper tackles the problem of training deep neural networks with noisy labels by proposing a Potential Energy based Mixture Model (PEMM), which incorporates data structure through a distance-based classifier with potential energy regularization, achieving state-of-the-art performance on real-world datasets.
Training deep neural networks (DNNs) from noisy labels is an important and challenging task. However, most existing approaches focus on the corrupted labels and ignore the importance of inherent data structure. To bridge the gap between noisy labels and data, inspired by the concept of potential energy in physics, we propose a novel Potential Energy based Mixture Model (PEMM) for noise-labels learning. We innovate a distance-based classifier with the potential energy regularization on its class centers. Embedding our proposed classifier with existing deep learning backbones, we can have robust networks with better feature representations. They can preserve intrinsic structures from the data, resulting in a superior noisy tolerance. We conducted extensive experiments to analyze the efficiency of our proposed model on several real-world datasets. Quantitative results show that it can achieve state-of-the-art performance.