Pruning for Better Domain Generalizability

TL;DR

This paper explores using pruning techniques, especially a novel method called DSS, to improve a model's ability to generalize across different domains, demonstrating significant empirical gains on benchmark datasets.

Contribution

The paper introduces DSS, a new pruning scoring method aimed at enhancing model robustness and domain generalization, and shows its effectiveness when combined with existing methods.

Findings

DSS improves domain generalization performance.

Pruning with DSS can be combined with state-of-the-art methods like MIRO.

Significant performance gains on MNIST-M and DomainBed benchmarks.

Abstract

In this paper, we investigate whether we could use pruning as a reliable method to boost the generalization ability of the model. We found that existing pruning method like L2 can already offer small improvement on the target domain performance. We further propose a novel pruning scoring method, called DSS, designed not to maintain source accuracy as typical pruning work, but to directly enhance the robustness of the model. We conduct empirical experiments to validate our method and demonstrate that it can be even combined with state-of-the-art generalization work like MIRO(Cha et al., 2022) to further boost the performance. On MNIST to MNIST-M, we could improve the baseline performance by over 5 points by introducing 60% channel sparsity into the model. On DomainBed benchmark and state-of-the-art MIRO, we can further boost its performance by 1 point only by introducing 10% sparsity into the model. Code can be found at: https://github.com/AlexSunNik/Pruning-for-Better-Domain-Generalizability