Fast Clifford Neural Layers
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912461119029248 |
|---|---|
| author | Xia, Tianxiang Neuwinger, Max Xiao, Lin |
| author_facet | Xia, Tianxiang Neuwinger, Max Xiao, Lin |
| contents | Clifford Neural Layers improve PDE modeling by introducing Clifford Algebra into neural networks. In this project we focus on optimizing the inference of 2/3D Clifford convolutional layers and multivector activation layers for one core CPU performance.
Overall, by testing on a real network block involving Clifford convolutional layers and multivector activation layers, we observe that our implementation is 30% faster than standard PyTorch implementation in relatively large data + network size (>L2 cache).
We open source our code base at https://github.com/egretwAlker/c-opt-clifford-layers |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_01040 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Fast Clifford Neural Layers Xia, Tianxiang Neuwinger, Max Xiao, Lin Machine Learning Artificial Intelligence Neural and Evolutionary Computing Performance Clifford Neural Layers improve PDE modeling by introducing Clifford Algebra into neural networks. In this project we focus on optimizing the inference of 2/3D Clifford convolutional layers and multivector activation layers for one core CPU performance. Overall, by testing on a real network block involving Clifford convolutional layers and multivector activation layers, we observe that our implementation is 30% faster than standard PyTorch implementation in relatively large data + network size (>L2 cache). We open source our code base at https://github.com/egretwAlker/c-opt-clifford-layers |
| title | Fast Clifford Neural Layers |
| topic | Machine Learning Artificial Intelligence Neural and Evolutionary Computing Performance |
| url | https://arxiv.org/abs/2507.01040 |