التفاصيل البيبلوغرافية
| العنوان: |
A Review of State-of-the-Art Mixed-Precision Neural Network Frameworks. |
| المؤلفون: |
Rakka M, Fouda ME, Khargonekar P, Kurdahi F |
| المصدر: |
IEEE transactions on pattern analysis and machine intelligence [IEEE Trans Pattern Anal Mach Intell] 2024 Apr 29; Vol. PP. Date of Electronic Publication: 2024 Apr 29. |
| Publication Model: |
Ahead of Print |
| نوع المنشور: |
Journal Article |
| اللغة: |
English |
| بيانات الدورية: |
Publisher: IEEE Computer Society Country of Publication: United States NLM ID: 9885960 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1939-3539 (Electronic) Linking ISSN: 00985589 NLM ISO Abbreviation: IEEE Trans Pattern Anal Mach Intell Subsets: MEDLINE |
| أسماء مطبوعة: |
Original Publication: [New York] IEEE Computer Society. |
| مستخلص: |
Mixed-precision Deep Neural Networks (DNNs) provide an efficient solution for hardware deployment, especially under resource constraints, while maintaining model accuracy. Identifying the ideal bit precision for each layer, however, remains a challenge given the vast array of models, datasets, and quantization schemes, leading to an expansive search space. Recent literature has addressed this challenge, resulting in several promising frameworks. This paper offers a comprehensive overview of the standard quantization classifications prevalent in existing studies. A detailed survey of current mixed-precision frameworks is provided, with an in-depth comparative analysis highlighting their respective merits and limitations. The paper concludes with insights into potential avenues for future research in this domain. |
| تواريخ الأحداث: |
Date Created: 20240429 Latest Revision: 20240502 |
| رمز التحديث: |
20240502 |
| DOI: |
10.1109/TPAMI.2024.3394390 |
| PMID: |
38683716 |
| قاعدة البيانات: |
MEDLINE |
Full Text Finder