Investigating the Impact of Weight Sharing Decisions on Knowledge Transfer in Continual Learning

TL;DR

This paper explores how different weight sharing strategies in continual learning affect forward knowledge transfer, demonstrating that task complexity and similarity influence optimal sharing decisions to improve accuracy.

Contribution

It provides an analysis of weight sharing decisions in continual learning, revealing how task characteristics guide optimal sharing for better transfer and accuracy.

Findings

Task complexity and similarity influence optimal weight sharing.

Sharing decisions based on findings improve task accuracy.

Structured subnetworks facilitate better knowledge transfer.

Abstract

Continual Learning (CL) has generated attention as a method of avoiding Catastrophic Forgetting (CF) in the sequential training of neural networks, improving network efficiency and adaptability to different tasks. Additionally, CL serves as an ideal setting for studying network behavior and Forward Knowledge Transfer (FKT) between tasks. Pruning methods for CL train subnetworks to handle the sequential tasks which allows us to take a structured approach to investigating FKT. Sharing prior subnetworks' weights leverages past knowledge for the current task through FKT. Understanding which weights to share is important as sharing all weights can yield sub-optimal accuracy. This paper investigates how different sharing decisions affect the FKT between tasks. Through this lens we demonstrate how task complexity and similarity influence the optimal weight sharing decisions, giving insights into the relationships between tasks and helping inform decision making in similar CL methods. We implement three sequential datasets designed to emphasize variation in task complexity and similarity, reporting results for both ResNet-18 and VGG-16. By sharing in accordance with the decisions supported by our findings, we show that we can improve task accuracy compared to other sharing decisions.