Learning From Multiple Experts: Self-paced Knowledge Distillation for Long-tailed Classification

ECCV 2020  ·  Liuyu Xiang, Guiguang Ding, Jungong Han ·

In real-world scenarios, data tends to exhibit a long-tailed distribution, which increases the difficulty of training deep networks. In this paper, we propose a novel self-paced knowledge distillation framework, termed Learning From Multiple Experts (LFME). Our method is inspired by the observation that networks trained on less imbalanced subsets of the distribution often yield better performances than their jointly-trained counterparts. We refer to these models as 'Experts', and the proposed LFME framework aggregates the knowledge from multiple 'Experts' to learn a unified student model. Specifically, the proposed framework involves two levels of adaptive learning schedules: Self-paced Expert Selection and Curriculum Instance Selection, so that the knowledge is adaptively transferred to the 'Student'. We conduct extensive experiments and demonstrate that our method is able to achieve superior performances compared to state-of-the-art methods. We also show that our method can be easily plugged into state-of-the-art long-tailed classification algorithms for further improvements.

PDF Abstract ECCV 2020 PDF ECCV 2020 Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Long-tail Learning ImageNet-LT LFME + OLTR Top-1 Accuracy 38.8 # 60
Long-tail Learning Places-LT LFME + OLTR Top-1 Accuracy 36.2 # 25