A Gated Residual Kolmogorov-Arnold Networks for Mixtures of Experts
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913610496737280 |
|---|---|
| author | Inzirillo, Hugo Genet, Remi |
| author_facet | Inzirillo, Hugo Genet, Remi |
| contents | This paper introduces KAMoE, a novel Mixture of Experts (MoE) framework based on Gated Residual Kolmogorov-Arnold Networks (GRKAN). We propose GRKAN as an alternative to the traditional gating function, aiming to enhance efficiency and interpretability in MoE modeling. Through extensive experiments on digital asset markets and real estate valuation, we demonstrate that KAMoE consistently outperforms traditional MoE architectures across various tasks and model types. Our results show that GRKAN exhibits superior performance compared to standard Gating Residual Networks, particularly in LSTM-based models for sequential tasks. We also provide insights into the trade-offs between model complexity and performance gains in MoE and KAMoE architectures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_15161 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A Gated Residual Kolmogorov-Arnold Networks for Mixtures of Experts Inzirillo, Hugo Genet, Remi Machine Learning Neural and Evolutionary Computing This paper introduces KAMoE, a novel Mixture of Experts (MoE) framework based on Gated Residual Kolmogorov-Arnold Networks (GRKAN). We propose GRKAN as an alternative to the traditional gating function, aiming to enhance efficiency and interpretability in MoE modeling. Through extensive experiments on digital asset markets and real estate valuation, we demonstrate that KAMoE consistently outperforms traditional MoE architectures across various tasks and model types. Our results show that GRKAN exhibits superior performance compared to standard Gating Residual Networks, particularly in LSTM-based models for sequential tasks. We also provide insights into the trade-offs between model complexity and performance gains in MoE and KAMoE architectures. |
| title | A Gated Residual Kolmogorov-Arnold Networks for Mixtures of Experts |
| topic | Machine Learning Neural and Evolutionary Computing |
| url | https://arxiv.org/abs/2409.15161 |