Gated Recurrent Neural Networks with Weighted Time-Delay Feedback
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2022
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913847141466112 |
|---|---|
| author | Erichson, N. Benjamin Lim, Soon Hoe Mahoney, Michael W. |
| author_facet | Erichson, N. Benjamin Lim, Soon Hoe Mahoney, Michael W. |
| contents | In this paper, we present a novel approach to modeling long-term dependencies in sequential data by introducing a gated recurrent unit (GRU) with a weighted time-delay feedback mechanism. Our proposed model, named $τ$-GRU, is a discretized version of a continuous-time formulation of a recurrent unit, where the dynamics are governed by delay differential equations (DDEs). We prove the existence and uniqueness of solutions for the continuous-time model and show that the proposed feedback mechanism can significantly improve the modeling of long-term dependencies. Our empirical results indicate that $τ$-GRU outperforms state-of-the-art recurrent units and gated recurrent architectures on a range of tasks, achieving faster convergence and better generalization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2212_00228 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Gated Recurrent Neural Networks with Weighted Time-Delay Feedback Erichson, N. Benjamin Lim, Soon Hoe Mahoney, Michael W. Machine Learning Neural and Evolutionary Computing In this paper, we present a novel approach to modeling long-term dependencies in sequential data by introducing a gated recurrent unit (GRU) with a weighted time-delay feedback mechanism. Our proposed model, named $τ$-GRU, is a discretized version of a continuous-time formulation of a recurrent unit, where the dynamics are governed by delay differential equations (DDEs). We prove the existence and uniqueness of solutions for the continuous-time model and show that the proposed feedback mechanism can significantly improve the modeling of long-term dependencies. Our empirical results indicate that $τ$-GRU outperforms state-of-the-art recurrent units and gated recurrent architectures on a range of tasks, achieving faster convergence and better generalization. |
| title | Gated Recurrent Neural Networks with Weighted Time-Delay Feedback |
| topic | Machine Learning Neural and Evolutionary Computing |
| url | https://arxiv.org/abs/2212.00228 |