A Topological Filter for Learning with Label Noise
A Topological Filter for Learning with Label Noise
复制标题
DOI：
--
复制
发表时间：
2020-12
期刊：
影响因子：
--
通讯作者：
Pengxiang Wu;Songzhu Zheng;Mayank Goswami;Dimitris N. Metaxas;Chao Chen
中科院分区：
文献类型：
--
作者：
Pengxiang Wu;Songzhu Zheng;Mayank Goswami;Dimitris N. Metaxas;Chao Chen
噪声标签会影响深度神经网络的性能。为了解决这一问题，本文提出了一种滤波标签噪声的新方法。与大多数依赖于噪声分类器后验概率的现有方法不同，我们专注于潜在表征空间中数据的更丰富的空间行为。通过利用数据的高阶拓扑信息，我们能够收集大部分干净的数据，训练出高质量的模型。从理论上证明了这种拓扑方法能够保证高概率地收集到干净的数据。实证结果表明，我们的方法优于最先进的技术，并且对广泛的噪声类型和水平具有鲁棒性。
Noisy labels can impair the performance of deep neural networks. To tackle this problem, in this paper, we propose a new method for filtering label noise. Unlike most existing methods relying on the posterior probability of a noisy classifier, we focus on the much richer spatial behavior of data in the latent representational space. By leveraging the high-order topological information of data, we are able to collect most of the clean data and train a high-quality model. Theoretically we prove that this topological approach is guaranteed to collect the clean data with high probability. Empirical results show that our method outperforms the state-of-the-arts and is robust to a broad spectrum of noise types and levels.