LTRL: Boosting Long-tail Recognition via Reflective Learning

TL;DR

This paper introduces reflecting learning, a new paradigm that improves long-tail recognition by reviewing past predictions, summarizing class relations, and correcting gradient conflicts, leading to state-of-the-art results.

Contribution

The paper proposes a novel reflecting learning approach that enhances long-tail recognition by integrating review, summarization, and correction processes, compatible with existing methods.

Findings

Achieves state-of-the-art performance on long-tail visual benchmarks.

Demonstrates the effectiveness of reflecting learning in handling imbalanced data.

Lightweight design allows easy integration with current methods.

Abstract

In real-world scenarios, where knowledge distributions exhibit long-tail. Humans manage to master knowledge uniformly across imbalanced distributions, a feat attributed to their diligent practices of reviewing, summarizing, and correcting errors. Motivated by this learning process, we propose a novel learning paradigm, called reflecting learning, in handling long-tail recognition. Our method integrates three processes for reviewing past predictions during training, summarizing and leveraging the feature relation across classes, and correcting gradient conflict for loss functions. These designs are lightweight enough to plug and play with existing long-tail learning methods, achieving state-of-the-art performance in popular long-tail visual benchmarks. The experimental results highlight the great potential of reflecting learning in dealing with long-tail recognition.