Austin Watkins

2 accepted papers

2024

Adversarially Robust Multi-task Representation Learning

Austin Watkins, Thanh Nguyen-Tang, Enayat Ullah, Raman Arora

NeurIPS 2024poster

We study adversarially robust transfer learning, wherein, given labeled data on multiple (source) tasks, the goal is to train a model with small robust error on a previously unseen (target) task. In particular, we consider a multi-task representation learning (MTRL) setting, i.e., we assume that the…

Cited by 1 Source ↗PDF ↗Scholar ↗

2023

Optimistic Rates for Multi-Task Representation Learning

Austin Watkins, Enayat Ullah, Thanh Nguyen-Tang, Raman Arora

NeurIPS 2023poster

We study the problem of transfer learning via Multi-Task Representation Learning (MTRL), wherein multiple source tasks are used to learn a good common representation, and a predictor is trained on top of it for the target task. Under standard regularity assumptions on the loss function and task dive…

Cited by 13 Source ↗PDF ↗Scholar ↗