Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2307.12713 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909225629777920 |
|---|---|
| author | Gauffriau, Adrien Silva, Iryna De Albuquerque Pagetti, Claire |
| author_facet | Gauffriau, Adrien Silva, Iryna De Albuquerque Pagetti, Claire |
| contents | Implementing deep neural networks in safety critical systems, in particular in the aeronautical domain, will require to offer adequate specification paradigms to preserve the semantics of the trained model on the final hardware platform. We propose to extend the nnef language in order to allow traceable distribution and parallelisation optimizations of a trained model. We show how such a specification can be implemented in cuda on a Xavier platform. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2307_12713 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Formal description of ML models for unambiguous implementation Gauffriau, Adrien Silva, Iryna De Albuquerque Pagetti, Claire Neural and Evolutionary Computing Implementing deep neural networks in safety critical systems, in particular in the aeronautical domain, will require to offer adequate specification paradigms to preserve the semantics of the trained model on the final hardware platform. We propose to extend the nnef language in order to allow traceable distribution and parallelisation optimizations of a trained model. We show how such a specification can be implemented in cuda on a Xavier platform. |
| title | Formal description of ML models for unambiguous implementation |
| topic | Neural and Evolutionary Computing |
| url | https://arxiv.org/abs/2307.12713 |