The Multiple Subnetwork Hypothesis: Enabling Multidomain Learning by Isolating Task-Specific Subnetworks in Feedforward Neural Networks

TL;DR

This paper introduces a method for isolating task-specific subnetworks within neural networks, enabling effective multitask learning without performance loss or catastrophic forgetting.

Contribution

It proposes a novel network structure and methodology that allows pruned networks to utilize unused weights for learning new tasks, enhancing multitask capabilities.

Findings

Networks can learn multiple tasks in parallel or sequence.

No performance degradation or catastrophic forgetting observed.

Method effective on benchmark datasets.

Abstract

Neural networks have seen an explosion of usage and research in the past decade, particularly within the domains of computer vision and natural language processing. However, only recently have advancements in neural networks yielded performance improvements beyond narrow applications and translated to expanded multitask models capable of generalizing across multiple data types and modalities. Simultaneously, it has been shown that neural networks are overparameterized to a high degree, and pruning techniques have proved capable of significantly reducing the number of active weights within the network while largely preserving performance. In this work, we identify a methodology and network representational structure which allows a pruned network to employ previously unused weights to learn subsequent tasks. We employ these methodologies on well-known benchmarking datasets for testing purposes and show that networks trained using our approaches are able to learn multiple tasks, which may be related or unrelated, in parallel or in sequence without sacrificing performance on any task or exhibiting catastrophic forgetting.