ZOQO: Zero-Order Quantized Optimization
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917890104492032 |
|---|---|
| author | Bar, Noga Giryes, Raja |
| author_facet | Bar, Noga Giryes, Raja |
| contents | The increasing computational and memory demands in deep learning present significant challenges, especially in resource-constrained environments. We introduce a zero-order quantized optimization (ZOQO) method designed for training models with quantized parameters and operations. Our approach leverages zero-order approximations of the gradient sign and adapts the learning process to maintain the parameters' quantization without the need for full-precision gradient calculations. We demonstrate the effectiveness of ZOQO through experiments in fine-tuning of large language models and black-box adversarial attacks. Despite the limitations of zero-order and quantized operations training, our method achieves competitive performance compared to full-precision methods, highlighting its potential for low-resource environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_06736 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ZOQO: Zero-Order Quantized Optimization Bar, Noga Giryes, Raja Machine Learning Computation and Language The increasing computational and memory demands in deep learning present significant challenges, especially in resource-constrained environments. We introduce a zero-order quantized optimization (ZOQO) method designed for training models with quantized parameters and operations. Our approach leverages zero-order approximations of the gradient sign and adapts the learning process to maintain the parameters' quantization without the need for full-precision gradient calculations. We demonstrate the effectiveness of ZOQO through experiments in fine-tuning of large language models and black-box adversarial attacks. Despite the limitations of zero-order and quantized operations training, our method achieves competitive performance compared to full-precision methods, highlighting its potential for low-resource environments. |
| title | ZOQO: Zero-Order Quantized Optimization |
| topic | Machine Learning Computation and Language |
| url | https://arxiv.org/abs/2501.06736 |