no code implementations • 16 Jun 2023 • Lifan Zhao, Shuming Kong, Yanyan Shen
To address this challenge, we propose DoubleAdapt, an end-to-end framework with two adapters, which can effectively adapt the data and the model to mitigate the effects of distribution shifts.
1 code implementation • ICLR 2022 • Shuming Kong, Yanyan Shen, Linpeng Huang
To achieve this, we use influence functions to estimate how relabeling a training sample would affect model's test performance and further develop a novel relabeling function R. We theoretically prove that applying R to relabel harmful training samples allows the model to achieve lower test loss than simply discarding them for any classification tasks using cross-entropy loss.