Search Results for author: Ammar Sherif

Found 1 papers, 1 papers with code

STG-MTL: Scalable Task Grouping for Multi-Task Learning Using Data Map

1 code implementation7 Jul 2023 Ammar Sherif, Abubakar Abid, Mustafa Elattar, Mohamed ElHelw

In our paper, we propose a new data-driven method that addresses these challenges and provides a scalable and modular solution for classification task grouping based on a re-proposed data-driven features, Data Maps, which capture the training dynamics for each classification task during the MTL training.

Multi-Task Learning

Cannot find the paper you are looking for? You can Submit a new open access paper.