Dynamic Curriculum Learning for Imbalanced Data Classification

TL;DR

This paper introduces Dynamic Curriculum Learning (DCL), a novel framework that adaptively manages data sampling and loss weighting to improve classification performance on imbalanced datasets in computer vision.

Contribution

The paper proposes a unified, adaptive curriculum learning framework with two schedulers for sampling and loss control, enhancing generalization on imbalanced data.

Findings

Achieved state-of-the-art results on CelebA and RAP datasets.

Effectively balances data distribution from easy to hard samples.

Improves classification accuracy in imbalanced attribute datasets.

Abstract

Human attribute analysis is a challenging task in the field of computer vision, since the data is largely imbalance-distributed. Common techniques such as re-sampling and cost-sensitive learning require prior-knowledge to train the system. To address this problem, we propose a unified framework called Dynamic Curriculum Learning (DCL) to online adaptively adjust the sampling strategy and loss learning in single batch, which resulting in better generalization and discrimination. Inspired by the curriculum learning, DCL consists of two level curriculum schedulers: (1) sampling scheduler not only manages the data distribution from imbalanced to balanced but also from easy to hard; (2) loss scheduler controls the learning importance between classification and metric learning loss. Learning from these two schedulers, we demonstrate our DCL framework with the new state-of-the-art performance on the widely used face attribute dataset CelebA and pedestrian attribute dataset RAP.