What is it about?
In Multi-Task Learning, tasks are commonly related considering all features in a symmetrical way. We have shown that estimating local transference (considering subsets of features) in an asymmetrical way reduces negative transference. Code on github.
Featured Image
Read the Original
This page is a summary of: Asymmetric Multi-Task Learning with Local Transference, ACM Transactions on Knowledge Discovery from Data, October 2022, ACM (Association for Computing Machinery),
DOI: 10.1145/3514252.
You can read the full text:
Contributors
The following have contributed to this page
