Some like it tough: Improving model generalization via progressively increasing the training difficulty

TL;DR

This paper introduces mini-batch trimming, a simple method that progressively increases training difficulty by focusing on harder samples, leading to improved generalization in neural network models.

Contribution

The paper proposes a novel, easy-to-integrate training strategy that enhances model generalization by emphasizing difficult samples during later training stages.

Findings

Improved test error across multiple image classification datasets.

Easy integration without modifying existing network architectures.

Enhanced model robustness through progressive difficulty increase.

Abstract

In this work, we propose to progressively increase the training difficulty during learning a neural network model via a novel strategy which we call mini-batch trimming. This strategy makes sure that the optimizer puts its focus in the later training stages on the more difficult samples, which we identify as the ones with the highest loss in the current mini-batch. The strategy is very easy to integrate into an existing training pipeline and does not necessitate a change of the network model. Experiments on several image classification problems show that mini-batch trimming is able to increase the generalization ability (measured via final test error) of the trained model.