Three-Way Trade-Off in Multi-Objective Learning: Optimization, Generalization and Conflict-Avoidance Article Swipe

MAKE IMPACT

Lisha Chen , Heshan Fernando , Yiming Ying , Tianyi Chen · YOU? · · 2023 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2305.20057

Multi-objective learning (MOL) problems often arise in emerging machine learning problems when there are multiple learning criteria, data modalities, or learning tasks. Different from single-objective learning, one of the critical challenges in MOL is the potential conflict among different objectives during the iterative optimization process. Recent works have developed various dynamic weighting algorithms for MOL such as MGDA and its variants, where the central idea is to find an update direction that avoids conflicts among objectives. Albeit its appealing intuition, empirical studies show that dynamic weighting methods may not always outperform static ones. To understand this theory-practical gap, we focus on a new stochastic variant of MGDA - the Multi-objective gradient with Double sampling (MoDo) algorithm, and study the generalization performance of the dynamic weighting-based MoDo and its interplay with optimization through the lens of algorithm stability. Perhaps surprisingly, we find that the key rationale behind MGDA -- updating along conflict-avoidant direction - may hinder dynamic weighting algorithms from achieving the optimal ${\cal O}(1/\sqrt{n})$ population risk, where $n$ is the number of training samples. We further demonstrate the impact of the variability of dynamic weights on the three-way trade-off among optimization, generalization, and conflict avoidance that is unique in MOL. We showcase the generality of our theoretical framework by analyzing other existing stochastic MOL algorithms under the framework. Experiments on various multi-task learning benchmarks are performed to demonstrate the practical applicability. Code is available at https://github.com/heshandevaka/Trade-Off-MOL.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2305.20057

PDF

https://arxiv.org/pdf/2305.20057

OA Status

green

Cited By

2

Related Works

10

OpenAlex ID

https://openalex.org/W4379089730

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4379089730

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2305.20057

Digital Object Identifier

Title

Three-Way Trade-Off in Multi-Objective Learning: Optimization, Generalization and Conflict-Avoidance

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2023

Year of publication

Publication date

2023-05-31

Full publication date if available

Authors

Lisha Chen, Heshan Fernando, Yiming Ying, Tianyi Chen

List of authors in order

Landing page

https://arxiv.org/abs/2305.20057

Publisher landing page

PDF URL

https://arxiv.org/pdf/2305.20057

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2305.20057

Direct OA link when available

Concepts

Weighting, Computer science, Machine learning, Generalization, Generality, Artificial intelligence, Stability (learning theory), Mathematical optimization, Algorithm, Mathematics, Psychology, Medicine, Psychotherapist, Mathematical analysis, Radiology

Top concepts (fields/topics) attached by OpenAlex

Cited by

2

Total citation count in OpenAlex

Citations by year (recent)

2025: 1, 2024: 1

Per-year citation counts (last 5 years)

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload