1 Hit in 2.1 sec

DLTTA: Dynamic Learning Rate for Test-time Adaptation on Cross-domain Medical Images [article]

Hongzheng Yang, Cheng Chen, Meirui Jiang, Quande Liu, Jianfeng Cao, Pheng Ann Heng, Qi Dou
2022 arXiv   pre-print
Test-time adaptation (TTA) has increasingly been an important topic to efficiently tackle the cross-domain distribution shift at test time for medical images from different institutions.  ...  To address this problem, we propose a novel dynamic learning rate adjustment method for test-time adaptation, called DLTTA, which dynamically modulates the amount of weights update for each test image  ...  Our main contributions are as follows: • We propose a new dynamic learning rate adjustment framework for test-time adaptation on cross-domain medical image analysis.  ... 
arXiv:2205.13723v1 fatcat:udwtqatuc5ef5pisq3qccvropi