Abstract | ||
---|---|---|
Machine learning is experiencing an explosion of software and hardware solutions, and needs industry-standard performance benchmarks to drive design and enable competitive evaluation. However, machine learning training presents a number of unique challenges to benchmarking that do not exist in other domains: (1) some optimizations that improve training throughput actually increase time to solution, (2) training is stochastic and time to solution has high variance, and (3) the software and hardware systems are so diverse that they cannot be fairly benchmarked with the same binary, code, or even hyperparameters. We present MLPerf, a machine learning benchmark that overcomes these challenges. We quantitatively evaluate the efficacy of MLPerf in driving community progress on performance and scalability across two rounds of results from multiple vendors. |
Year | Venue | DocType |
---|---|---|
2020 | MLSys | Conference |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
33 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mattson Peter | 1 | 0 | 0.34 |
Cheng Christine | 2 | 0 | 0.34 |
Coleman Cody | 3 | 0 | 0.34 |
Gregory Frederick Diamos | 4 | 1117 | 51.07 |
Micikevicius Paulius | 5 | 0 | 0.34 |
David A. Patterson | 6 | 11093 | 1925.05 |
Hanlin Tang | 7 | 29 | 5.46 |
Gu-Yeon Wei | 8 | 1927 | 214.15 |
Peter Bailis | 9 | 563 | 49.89 |
Bittorf Victor | 10 | 0 | 0.34 |
David Brooks | 11 | 5518 | 422.08 |
Chen Dehao | 12 | 0 | 0.34 |
Debojyoti Dutta | 13 | 299 | 27.48 |
Udit Gupta | 14 | 74 | 6.27 |
Kim M. Hazelwood | 15 | 2465 | 110.46 |
Hock Andrew | 16 | 0 | 0.34 |
Huang Xinyuan | 17 | 0 | 0.34 |
Jia Bill | 18 | 0 | 0.34 |
Daniel D. Kang | 19 | 4 | 5.23 |
Kanter David | 20 | 0 | 0.34 |
Kumar Naveen | 21 | 0 | 0.34 |
Liao Jeffery | 22 | 0 | 0.34 |
Deepak Narayanan | 23 | 47 | 7.42 |
Tayo Oguntebi | 24 | 360 | 13.47 |
Pekhimenko Gennady | 25 | 0 | 0.34 |
Pentecost Lillian | 26 | 0 | 0.34 |
Vijay Janapa Reddi | 27 | 2931 | 140.26 |
Robie Taylor | 28 | 0 | 0.34 |
John Tom St. | 29 | 0 | 0.34 |
Wu Carole-Jean | 30 | 0 | 0.34 |
Xu Lingjie | 31 | 0 | 0.34 |
Matei Zaharia | 32 | 9101 | 407.89 |
Zaharia Matei | 33 | 0 | 0.34 |