Dragonfly: a modular deep reinforcement learning library
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908469030813696 |
|---|---|
| author | Viquerat, Jonathan Garnier, Paul Bateni, Amirhossein Hachem, Elie |
| author_facet | Viquerat, Jonathan Garnier, Paul Bateni, Amirhossein Hachem, Elie |
| contents | Dragonfly is a deep reinforcement learning library focused on modularity, in order to ease experimentation and developments. It relies on a json serialization that allows to swap building blocks and perform parameter sweep, while minimizing code maintenance. Some of its features are specifically designed for CPU-intensive environments, such as numerical simulations. Its performance on standard agents using common benchmarks compares favorably with the literature. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_03778 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Dragonfly: a modular deep reinforcement learning library Viquerat, Jonathan Garnier, Paul Bateni, Amirhossein Hachem, Elie Machine Learning Dragonfly is a deep reinforcement learning library focused on modularity, in order to ease experimentation and developments. It relies on a json serialization that allows to swap building blocks and perform parameter sweep, while minimizing code maintenance. Some of its features are specifically designed for CPU-intensive environments, such as numerical simulations. Its performance on standard agents using common benchmarks compares favorably with the literature. |
| title | Dragonfly: a modular deep reinforcement learning library |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2505.03778 |