Disentangling Transfer and Interference in Multi-Domain Learning

Abstract

Humans are incredibly good at transferring knowledge from one domain to another, enabling rapid learning of new tasks. Likewise, transfer learning has enabled enormous success in many computer vision problems using pretraining. However, the benefits of transfer in multi-domain learning, where a network learns multiple tasks defined by different datasets, has not been adequately studied. Learning multiple domains could be beneficial, or these domains could interfere with each other given limited network capacity. Understanding how deep neural networks of varied capacity facilitate transfer across inputs from different distributions is a critical step towards open world learning. In this work, we decipher the conditions where interference and knowledge transfer occur in multi-domain learning. We propose new metrics disentangling interference and transfer, set up experimental protocols, and examine the roles of network capacity, task grouping, and dynamic loss weighting in reducing interference and facilitating transfer.

AAAI 2022 PracticalDL Workshop

References in corpus (7)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Three scenarios for continual learning
• Similarity of Neural Network Representations Revisited
• A Comprehensive Survey on Transfer Learning
• A Comprehensive Study of Class Incremental Learning Algorithms for Visual Tasks
• Understanding and Improving Information Transfer in Multi-Task Learning
• Scaling Laws for Transfer