P-DIFF: Learning Classifier with Noisy Labels based on Probability Difference Distributions
| Autor: | Wei Hu, QiHao Zhao, Yangyu Huang, Fan Zhang |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
FOS: Computer and information sciences
Artificial neural network Noise measurement Computer science business.industry Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition Pattern recognition Sample (statistics) 02 engineering and technology 010501 environmental sciences Overfitting 01 natural sciences Classifier (linguistics) Pattern recognition (psychology) 0202 electrical engineering electronic engineering information engineering Benchmark (computing) Task analysis 020201 artificial intelligence & image processing Artificial intelligence business 0105 earth and related environmental sciences |
| Zdroj: | ICPR |
| DOI: | 10.48550/arxiv.2009.06382 |
| Popis: | Learning deep neural network (DNN) classifier with noisy labels is a challenging task because the DNN can easily over-fit on these noisy labels due to its high capability. In this paper, we present a very simple but effective training paradigm called P-DIFF, which can train DNN classifiers but obviously alleviate the adverse impact of noisy labels. Our proposed probability difference distribution implicitly reflects the probability of a training sample to be clean, then this probability is employed to re-weight the corresponding sample during the training process. P-DIFF can also achieve good performance even without prior knowledge on the noise rate of training samples. Experiments on benchmark datasets also demonstrate that P-DIFF is superior to the state-of-the-art sample selection methods. Comment: ICPR 2020. Codes are available at https://github.com/fistyee/P-DIFF |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|