Mutual Learning for Finetuning Click-Through Rate Prediction Models

TL;DR

This paper explores mutual learning as a method to enhance click-through rate prediction models, demonstrating that models can mutually improve each other and achieve up to 0.66% relative performance gains on benchmark datasets.

Contribution

It introduces the application of mutual learning between equal models for CTR prediction, showing improvements over traditional knowledge distillation methods.

Findings

Mutual learning improved CTR prediction accuracy by up to 0.66%.

The approach was validated on Criteo and Avazu datasets.

Mutual learning benefits models with similar complexity.

Abstract

Click-Through Rate (CTR) prediction has become an essential task in digital industries, such as digital advertising or online shopping. Many deep learning-based methods have been implemented and have become state-of-the-art models in the domain. To further improve the performance of CTR models, Knowledge Distillation based approaches have been widely used. However, most of the current CTR prediction models do not have much complex architectures, so it's hard to call one of them 'cumbersome' and the other one 'tiny'. On the other hand, the performance gap is also not very large between complex and simple models. So, distilling knowledge from one model to the other could not be worth the effort. Under these considerations, Mutual Learning could be a better approach, since all the models could be improved mutually. In this paper, we showed how useful the mutual learning algorithm could be when it is between equals. In our experiments on the Criteo and Avazu datasets, the mutual learning algorithm improved the performance of the model by up to 0.66% relative improvement.