Disentangling Transfer and Interference in Multi-Domain Learning

TL;DR

This paper investigates how neural networks transfer knowledge and interfere across multiple domains, proposing metrics and protocols to understand the effects of capacity, task grouping, and loss weighting on multi-domain learning.

Contribution

It introduces new metrics and experimental protocols to disentangle transfer and interference, analyzing their dependence on network capacity and task management strategies.

Findings

Network capacity influences transfer and interference balance.

Task grouping and dynamic loss weighting can reduce interference.

Properly managed multi-domain learning enhances transfer benefits.

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.