Mini-batch trimming
Mini-batch trimming - a curriculum learning method for improving model generalization via progressively increasing the training difficulty
The code repository contains Python scripts demonstrating "mini-batch trimming" strategy as proposed in paper https://arxiv.org/abs/2110.13058. This curriculum learning strategy progressively increasing training difficulty, which leads to a better generalization of the trained model.
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.