paper-with-me

홈 › Papers

Dual-Balancing for Multi-Task Learning

2023-08-23 · Baijiong Lin, Weisen Jiang, Feiyang Ye, Yu Zhang, Pengguang Chen, Ying-Cong Chen, Shu Liu, James T. Kwok

Multi-task learning (MTL), a learning paradigm to learn multiple related tasks simultaneously, has achieved great success in various fields. However, task balancing problem remains a significant challenge in MTL, with the disparity in loss/gradient scales often leading to performance compromises. In this paper, we propose a Dual-Balancing Multi-Task Learning (DB-MTL) method to alleviate the task balancing problem from both loss and gradient perspectives. Specifically, DB-MTL ensures loss-scale balancing by performing a logarithm transformation on each task loss, and guarantees gradient-magnitude balancing via normalizing all task gradients to the same magnitude as the maximum gradient norm. Extensive experiments conducted on several benchmark datasets consistently demonstrate the state-of-the-art performance of DB-MTL.

📄 PDF Abstract BibTeX arXiv:2308.12029

Code (1)

median-research-group/libmtl pytorch

Tasks

Multi-Task Learning

Similar Papers 제목 키워드 기반

Dual-Balancing for Physics-Informed Neural Networks

2025-05-16 · Chenhong Zhou, Jie Chen, Zaifeng Yang, Ching Eng Png

Physics-informed neural networks (PINNs) have emerged as a new learning paradigm for solving partial differential equations (PDEs) by enforcing the constraints of physical equations, boundary conditions (BCs), and initia…

Parameter Competition Balancing for Model Merging

2024-10-03 · Guodong Du, Junlin Lee, Jing Li, Runhua Jiang 외

While fine-tuning pretrained models has become common practice, these models often underperform outside their specific domains. Recently developed model merging techniques enable the direct integration of multiple models…

Domain Generalizationmodel

Learning Multiple Pixelwise Tasks Based on Loss Scale Balancing

2021-01-01 · ICCV 2021 10 · Jae-Han Lee, Chul Lee, Chang-Su Kim

We propose a novel loss weighting algorithm, called loss scale balancing (LSB), for multi-task learning (MTL) of pixelwise vision tasks. An MTL model is trained to estimate multiple pixelwise predictions using an ove…

Multi-Task Learning

FreeBalance: Pre-Routing Online Moe Load Balancing via Residual Workload Prediction

2026-08-14 · Pengfei Chen, Yize Wu, Shouxu Kuang, Ke Gao 외 arxiv

Load imbalance poses a major bottleneck to the efficiency of expert parallelism in distributed inference of Mixture-of-Experts (MoE) models. The most heavily loaded rank stalls global execution due to skewed routing dist…

Achievement-Based Training Progress Balancing for Multi-Task Learning

2023-01-01 · ICCV 2023 1 · Hayoung Yun, Hanjoo Cho

Multi-task learning faces two challenging issues: (1) the high cost of annotating labels for all tasks and (2) balancing the training progress of various tasks with different natures. To resolve the label annotation …

Depth EstimationMulti-Task Learningobject-detectionObject Detection+1