2020 IEEE Hot Chips 32 Symposium (HCS) 2020
DOI: 10.1109/hcs49909.2020.9220687
|View full text |Cite
|
Sign up to set email alerts
|

Compute substrate for Software 2.0

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1

Citation Types

0
1
0

Year Published

2024
2024
2024
2024

Publication Types

Select...
1

Relationship

0
1

Authors

Journals

citations
Cited by 1 publication
(1 citation statement)
references
References 0 publications
0
1
0
Order By: Relevance
“…However, choosing an appropriate memory specification is always a challenge. In Figure 2, we surveyed 16 popular industrial neural network processors with various memory/performance/area characteristics, where nine of them target the training domain [6,11,24,34,35,40,41,48,60,63,69] and seven target model inference [1, 7, 8, 26-28, 49, 65]. According to the survey, we can observe several trends as follows:…”
Section: Design Of Neural Network Acceleratorsmentioning
confidence: 99%
“…However, choosing an appropriate memory specification is always a challenge. In Figure 2, we surveyed 16 popular industrial neural network processors with various memory/performance/area characteristics, where nine of them target the training domain [6,11,24,34,35,40,41,48,60,63,69] and seven target model inference [1, 7, 8, 26-28, 49, 65]. According to the survey, we can observe several trends as follows:…”
Section: Design Of Neural Network Acceleratorsmentioning
confidence: 99%