Can I read UM-Adapt: Unsupervised Multi-Task Adaptation Using Adversarial Cross-Task Distillation on EtoBox?

UM-Adapt: Unsupervised Multi-Task Adaptation Using Adversarial Cross-Task Distillation by Kundu, Jogendra Nath; Lakkakula, Nishank; Babu, R. Venkatesh is a scholarly article available to read on EtoBox.

What is UM-Adapt: Unsupervised Multi-Task Adaptation Using Adversarial Cross-Task Distillation about?

Aiming towards human-level generalization, there is a need to explore adaptable representation learning methods with greater transferability. Most existing approaches independently address task-transferability and cross-domain adaptation, resulting in limited generalization. In this paper, we propose UM-Adapt - a unified framework to effectively perform unsupervised domain adaptation for spatially-structured prediction tasks, simultaneously maintaining a balanced performance across individual tasks in a multi-task setting. To realize this, we propose two novel regularization strategies; a) Contour-based content regularization (CCR) and b) exploitation of inter-task coherency using a cross-task distillation module. Furthermore, avoiding a conventional ad-hoc domain discriminator, we re-utilize the cross-task distillation loss as output of an energy function to adversarially minimize the input domain discrepancy. Through extensive experiments, we demonstrate superior generalizability of the learned representations simultaneously for multiple tasks under domain-shifts from synthetic to natural environments. UM-Adapt yields state-of-the-art transfer learning results on ImageNet classif

Author

Kundu, Jogendra Nath; Lakkakula, Nishank; Babu, R. Venkatesh

Published

2019

Language

EN