Revisiting scalarization in multi-task learning: A theoretical perspective

Y Hu, R ** for multi-constraint safe reinforcement learning
Y Yao, Z Liu, Z Cen, P Huang… - … Annual Learning for …, 2024 - proceedings.mlr.press
Online safe reinforcement learning (RL) involves training a policy that maximizes task
efficiency while satisfying constraints via interacting with the environments. In this paper, our …

Min-max multi-objective bilevel optimization with applications in robust machine learning

A Gu, S Lu, P Ram, TW Weng - The Eleventh International …, 2022 - openreview.net
We consider a generic min-max multi-objective bilevel optimization problem with
applications in robust machine learning such as representation learning and …

Smooth Tchebycheff Scalarization for Multi-Objective Optimization

X Lin, X Zhang, Z Yang, F Liu, Z Wang… - arxiv preprint arxiv …, 2024 - arxiv.org
Multi-objective optimization problems can be found in many real-world applications, where
the objectives often conflict each other and cannot be optimized by a single solution. In the …

Fair Resource Allocation in Multi-Task Learning

H Ban, K Ji - arxiv preprint arxiv:2402.15638, 2024 - arxiv.org
By jointly learning multiple tasks, multi-task learning (MTL) can leverage the shared
knowledge across tasks, resulting in improved data efficiency and generalization …

Challenging Common Assumptions in Multi-task Learning

C Elich, L Kirchdorfer, JM Köhler, L Schott - arxiv preprint arxiv …, 2023 - arxiv.org
While multi-task learning (MTL) has gained significant attention in recent years, its
underlying mechanisms remain poorly understood. Recent methods did not yield consistent …

PSMGD: Periodic Stochastic Multi-Gradient Descent for Fast Multi-Objective Optimization

M Xu, P Ju, J Liu, H Yang - arxiv preprint arxiv:2412.10961, 2024 - arxiv.org
Multi-objective optimization (MOO) lies at the core of many machine learning (ML)
applications that involve multiple, potentially conflicting objectives (eg, multi-task learning …