Aligned Multi Objective Optimization

TL;DR

This paper introduces the Aligned Multi-Objective Optimization framework, addressing scenarios with related tasks that improve multiple objectives simultaneously, and proposes algorithms with theoretical guarantees for better performance.

Contribution

It presents a novel framework and algorithms for optimizing multiple related objectives without conflicts, filling a gap in existing multi-objective optimization methods.

Findings

Algorithms outperform naive approaches in theoretical guarantees

Framework effectively handles large numbers of related objectives

Addresses a gap in optimization for multi-task and multi-objective scenarios

Abstract

To date, the multi-objective optimization literature has mainly focused on conflicting objectives, studying the Pareto front, or requiring users to balance tradeoffs. Yet, in machine learning practice, there are many scenarios where such conflict does not take place. Recent findings from multi-task learning, reinforcement learning, and LLMs training show that diverse related tasks can enhance performance across objectives simultaneously. Despite this evidence, such phenomenon has not been examined from an optimization perspective. This leads to a lack of generic gradient-based methods that can scale to scenarios with a large number of related objectives. To address this gap, we introduce the Aligned Multi-Objective Optimization framework, propose new algorithms for this setting, and provide theoretical guarantees of their superior performance compared to naive approaches.