Learning More Generalized Experts by Merging Experts in Mixture-of-Experts

TL;DR

This paper proposes merging frequently used experts in a mixture-of-experts model to learn more generalized features, improving transfer learning and reducing catastrophic forgetting in multi-domain tasks.

Contribution

It introduces a novel expert merging strategy based on usage frequency to enhance generalization and continual learning in mixture-of-experts models.

Findings

Merging experts improves model generalization.

The approach mitigates catastrophic forgetting.

Enhanced transfer learning performance.

Abstract

We observe that incorporating a shared layer in a mixture-of-experts can lead to performance degradation. This leads us to hypothesize that learning shared features poses challenges in deep learning, potentially caused by the same feature being learned as various different features. To address this issue, we track each expert's usage frequency and merge the two most frequently selected experts. We then update the least frequently selected expert using the combination of experts. This approach, combined with the subsequent learning of the router's expert selection, allows the model to determine if the most frequently selected experts have learned the same feature differently. If they have, the combined expert can be further trained to learn a more general feature. Consequently, our algorithm enhances transfer learning and mitigates catastrophic forgetting when applied to multi-domain task incremental learning.