Minerva: Enabling low-power, highly-accurate deep neural network accelerators B Reagen, P Whatmough, R Adolf, S Rama, H Lee, SK Lee, ... ACM SIGARCH Computer Architecture News 44 (3), 267-278, 2016 | 745 | 2016 |
14.3 A 28nm SoC with a 1.2 GHz 568nJ/prediction sparse deep-neural-network engine with> 0.1 timing error rate tolerance for IoT applications PN Whatmough, SK Lee, H Lee, S Rama, D Brooks, GY Wei 2017 IEEE International Solid-State Circuits Conference (ISSCC), 242-243, 2017 | 207 | 2017 |
Fathom: Reference workloads for modern deep learning methods R Adolf, S Rama, B Reagen, GY Wei, D Brooks 2016 IEEE International Symposium on Workload Characterization (IISWC), 1-10, 2016 | 202 | 2016 |
Early dse and automatic generation of coarse-grained merged accelerators I Brumar, G Zacharopoulos, Y Yao, S Rama, D Brooks, GY Wei ACM Transactions on Embedded Computing Systems 22 (2), 1-29, 2023 | 10 | 2023 |
Application of approximate matrix multiplication to neural networks and distributed SLAM B Plancher, CD Brumar, I Brumar, L Pentecost, S Rama, D Brooks 2019 IEEE High Performance Extreme Computing Conference (HPEC), 1-7, 2019 | 7 | 2019 |
Methods of communication avoidance in parallel solutions of partial differential equations LS White, G Dasika, SV Rama US Patent App. 17/561,227, 2023 | | 2023 |
Machine learning based stabilizer for numerical methods SV Rama, G Dasika, LS White US Patent App. 17/550,882, 2023 | | 2023 |
The Design and Evolution of Deep Learning Workloads R Adolf, S Rama, B Reagen, GY Wei, D Brooks IEEE MICRO 37 (1), 18-21, 2017 | | 2017 |
Using neural networks to reduce communication in numerical solution of partial differential equations L White, G Dasika, S Rama | | |
2 Counting Distinct Elements in a Stream JN Scribe, S Rama | | |
2 Load Balancing JN Scribe, S Rama | | |
ISSCC 2017/SESSION 14/DEEP-LEARNING PROCESSORS/14.3 PN Whatmough, SK Lee, H Lee, S Rama, D Brooks, GY Wei | | |