Results 71 to 80 of about 3,537 (211)

Rep-MTL: Unleashing the Power of Representation-Level Task Saliency for Multi-Task Learning

2025 IEEE/CVF International Conference on Computer Vision (ICCV)
Despite the promise of Multi-Task Learning in leveraging complementary knowledge across tasks, existing multi-task optimization (MTO) techniques remain fixated on resolving conflicts via optimizer-centric loss scaling and gradient manipulation strategies, yet fail to deliver consistent gains. In this paper, we argue that the shared representation space,
Zedong Wang, Siyuan Li, Dan Xu
openaire   +2 more sources

Validating Thermodynamic Models of Arc‐Magma Differentiation and Training Neural Networks for Rapid Thermodynamic Property Inference

Geochemistry, Geophysics, Geosystems, Volume 27, Issue 4, April 2026.
Abstract Magmatic systems are the products of the migration and storage of compositionally evolving magmas within the solid crust. Direct computation of thermodynamic properties during magma transport remains a major challenge in multiphase‐transport modeling due to its computational cost.
Lorenzo G. Candioti, Chetan L. Nathwani, Cyril Chelle‐Michou   +2 more
wiley   +1 more source

Multi‐Task Learning for Airport Surface Surveillance: A Review

Expert Systems, Volume 43, Issue 4, April 2026.
ABSTRACT The rapid growth of air transportation has surpassed the capabilities of traditional airport surveillance methods, such as visual observation and auxiliary equipment (e.g., ADS‐B, MLAT, radar), which struggle to provide all‐area, all‐weather situation awareness.
Daoyong Fu, Xiangtong Wang, Fangrui Wu, Songchen Han, Binbin Liang, Wei Li, Ke Yang   +6 more
wiley   +1 more source

Leveraging multi-task learning regressor chains for small and sparse tabular data in materials design

Machine Learning: Science and Technology
Machine learning has become increasingly important in materials design, yet traditional single-task learning (STL) models fail to fully exploit the potential of available data in scenarios involving multiple targets and incomplete datasets.
Felix Conrad, Hajo Wiemer, Steffen Ihlenfeldt   +2 more
doaj   +1 more source

Generative Multi-Task Learning for Text Classification

IEEE Access, 2020
Multi-task learning leverages potential correlations among related tasks to extract common features and yield performance gains. In this paper, a generative multi-task learning (MTL) approach for text classification and categorization is proposed, which ...
Wei Zhao, Hui Gao, Shuhui Chen, Nan Wang
doaj   +1 more source

Pangenome Analysis Reveals Structural Variations Associated With Citric Acid Accumulation in Prunus mume

Plant Biotechnology Journal, Volume 24, Issue 4, Page 2623-2641, April 2026.
ABSTRACT Pangenome can reveal a large number of variations, providing a more comprehensive view of the genetic diversity of species that a single reference genome cannot surpass. Here, we assembled the haplotype telomere‐to‐telomere genome and 10 chromosome‐level genomes, integrated with two previously reported genomes, and constructed a graph ...
Xiao Huang, Ximeng Lin, Pengyu Zhou, Wei Tan, Feng Gao, Zhaojun Ni, Ting Shi, Yang Bai, Chengdong Ma, Yufan Ma, Minglu Li, Faisal Hayat, Qixiang Zhang, Ryutaro Tao, Jianmin Wan, Zhihong Gao   +15 more
wiley   +1 more source

Multi-task gradient descent for multi-task learning

, 2020
Multi-Task Learning (MTL) aims to simultaneously solve a group of related learning tasks by leveraging the salutary knowledge memes contained in the multiple tasks to improve the generalization performance.
Gupta, Abhishek, Bai, Lu, Ong, Yew-Soon, He, Tiantian   +3 more
core   +1 more source

Learning multi-level task groups in multi-task learning

, 2015
In multi-task learning (MTL), multiple related tasks are learned jointly by sharing information across them. Many MTL algorithms have been proposed to learn the underlying task groups. However, those methods are limited to learn the task groups at only a
Han, Lei, Zhang, Yu
core   +2 more sources

Automating multi-task learning on optical neural networks with weight sharing and physical rotation

Scientific Reports
The democratization of AI encourages multi-task learning (MTL), demanding more parameters and processing time. To achieve highly energy-efficient MTL, Diffractive Optical Neural Networks (DONNs) have garnered attention due to extremely low energy and ...
Shanglin Zhou, Yingjie Li, Weilu Gao, Cunxi Yu, Caiwen Ding   +4 more
doaj   +1 more source

Multi-task deep learning for simultaneous prediction of steel purity and carbon capture rate using membrane separation technology in integrated steelmaking processes

Array
Steel production significantly contributes to global CO2 emissions, demanding simultaneous optimization of product quality and environmental performance.
Somboon Sukpancharoen, Pakon Sakdee, Natacha Phetyim, Rinlada Sirisangsawang, Chayut Sungsook   +4 more
doaj   +1 more source