Adaptive Data Fusion for Multi-task Non-smooth Optimization

TL;DR

This paper introduces an adaptive data fusion method for multi-task non-smooth optimization, enhancing sample efficiency by exploiting commonalities among objectives, with proven statistical guarantees and demonstrated advantages in experiments.

Contribution

It presents a novel adaptive data fusion technique that leverages shared structures in multi-task non-smooth optimization, providing theoretical guarantees and empirical improvements.

Findings

Significant sample efficiency gains over benchmarks

Theoretical statistical guarantees established

Effective on both synthetic and real datasets

Abstract

We study the problem of multi-task non-smooth optimization that arises ubiquitously in statistical learning, decision-making and risk management. We develop a data fusion approach that adaptively leverages commonalities among a large number of objectives to improve sample efficiency while tackling their unknown heterogeneities. We provide sharp statistical guarantees for our approach. Numerical experiments on both synthetic and real data demonstrate significant advantages of our approach over benchmarks.