Noise-Resistant Deep Metric Learning with Probabilistic Instance Filtering
arXiv:2108.01431
Abstract
Noisy labels are commonly found in real-world data, which cause performance degradation of deep neural networks. Cleaning data manually is labour-intensive and time-consuming. Previous research mostly focuses on enhancing classification models against noisy labels, while the robustness of deep metric learning (DML) against noisy labels remains less well-explored. In this paper, we bridge this important gap by proposing Probabilistic Ranking-based Instance Selection with Memory (PRISM) approach for DML. PRISM calculates the probability of a label being clean, and filters out potentially noisy samples. Specifically, we propose a novel method, namely the von Mises-Fisher Distribution Similarity (vMF-Sim), to calculate this probability by estimating a von Mises-Fisher (vMF) distribution for each data class. Compared with the existing average similarity method (AvgSim), vMF-Sim considers the variance of each class in addition to the average similarity. With such a design, the proposed approach can deal with challenging DML situations in which the majority of the samples are noisy. Extensive experiments on both synthetic and real-world noisy dataset show that the proposed approach achieves up to 8.37% higher Precision@1 compared with the best performing state-of-the-art baseline approaches, within reasonable training time.
arXiv commentUnder review. Journal version of arXiv:2103.16047
References in corpus (9)
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Bootstrap your own latent: A new approach to self-supervised Learning
- In Defense of the Triplet Loss for Person Re-Identification
- DivideMix: Learning with Noisy Labels as Semi-supervised Learning
- Beyond Synthetic Noise: Deep Learning on Controlled Noisy Labels
- Distance-based Confidence Score for Neural Network Classifiers
- von Mises-Fisher Mixture Model-based Deep learning: Application to Face Verification
- Large-scale Landmark Retrieval/Recognition under a Noisy and Diverse Dataset
- PyTorch Metric Learning