A Metric Driven Approach to Mixed Precision Training
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911979550015488 |
|---|---|
| author | Rasquinha, Mitchelle Tabak, Gil |
| author_facet | Rasquinha, Mitchelle Tabak, Gil |
| contents | As deep learning methodologies have developed, it has been generally agreed that increasing neural network size improves model quality. However, this is at the expense of memory and compute requirements, which also need to be increased. Various efficiency techniques have been proposed to rein in hardware costs, one being the use of low precision numerics. Recent accelerators have introduced several different 8-bit data types to help accommodate DNNs in terms of numerics. In this paper, we identify a metric driven methodology to aid in the choice of numerics. We demonstrate how such a methodology can help scale training of a language representation model. The technique can be generalized to other model architectures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_02897 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A Metric Driven Approach to Mixed Precision Training Rasquinha, Mitchelle Tabak, Gil Machine Learning Artificial Intelligence As deep learning methodologies have developed, it has been generally agreed that increasing neural network size improves model quality. However, this is at the expense of memory and compute requirements, which also need to be increased. Various efficiency techniques have been proposed to rein in hardware costs, one being the use of low precision numerics. Recent accelerators have introduced several different 8-bit data types to help accommodate DNNs in terms of numerics. In this paper, we identify a metric driven methodology to aid in the choice of numerics. We demonstrate how such a methodology can help scale training of a language representation model. The technique can be generalized to other model architectures. |
| title | A Metric Driven Approach to Mixed Precision Training |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2408.02897 |