Learning to Auto Weight: Entirely Data-driven and Highly Efficient Weighting Framework

TL;DR

This paper introduces LAW, a data-driven, efficient framework for adaptive example weighting that improves training accuracy on biased datasets without prior knowledge or extensive hyperparameter tuning.

Contribution

The paper presents a novel, fully data-driven weighting framework that adaptively finds optimal weighting policies during training, outperforming traditional methods.

Findings

LAW achieves higher accuracy on biased CIFAR and ImageNet datasets.

The framework outperforms standard training pipelines and baseline methods.

It effectively reduces training bias through adaptive, step-dependent weighting policies.

Abstract

Example weighting algorithm is an effective solution to the training bias problem, however, most previous typical methods are usually limited to human knowledge and require laborious tuning of hyperparameters. In this paper, we propose a novel example weighting framework called Learning to Auto Weight (LAW). The proposed framework finds step-dependent weighting policies adaptively, and can be jointly trained with target networks without any assumptions or prior knowledge about the dataset. It consists of three key components: Stage-based Searching Strategy (3SM) is adopted to shrink the huge searching space in a complete training process; Duplicate Network Reward (DNR) gives more accurate supervision by removing randomness during the searching process; Full Data Update (FDU) further improves the updating efficiency. Experimental results demonstrate the superiority of weighting policy explored by LAW over standard training pipeline. Compared with baselines, LAW can find a better weighting schedule which achieves much more superior accuracy on both biased CIFAR and ImageNet.