Collaborative Prediction: To Join or To Disjoin Datasets

TL;DR

This paper investigates algorithms for selecting and merging datasets from various sources to improve predictive model performance, providing theoretical guarantees and demonstrating effectiveness in linear regression and machine learning tasks.

Contribution

It introduces a practical algorithm with theoretical guarantees for dataset selection and merging to minimize population loss in prediction models.

Findings

Algorithm reduces population loss with high probability

Effective in linear regression and broader machine learning applications

Code implementation available online

Abstract

With the recent rise of generative Artificial Intelligence (AI), the need of selecting high-quality dataset to improve machine learning models has garnered increasing attention. However, some part of this topic remains underexplored, even for simple prediction models. In this work, we study the problem of developing practical algorithms that select appropriate dataset to minimize population loss of our prediction model with high probability. Broadly speaking, we investigate when datasets from different sources can be effectively merged to enhance the predictive model's performance, and propose a practical algorithm with theoretical guarantees. By leveraging an oracle inequality and data-driven estimators, the algorithm reduces population loss with high probability. Numerical experiments demonstrate its effectiveness in both standard linear regression and broader machine learning applications. Code is available at https://github.com/kkrokii/collaborative_prediction.