Paper Info
Title
MLPerf Training Benchmark
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 Peter100.34
Cheng Christine200.34
Coleman Cody300.34
Gregory Frederick Diamos4111751.07
Micikevicius Paulius500.34
David A. Patterson6110931925.05
Hanlin Tang7295.46
Gu-Yeon Wei81927214.15
Peter Bailis956349.89
Bittorf Victor1000.34
David Brooks115518422.08
Chen Dehao1200.34
Debojyoti Dutta1329927.48
Udit Gupta14746.27
Kim M. Hazelwood152465110.46
Hock Andrew1600.34
Huang Xinyuan1700.34
Jia Bill1800.34
Daniel D. Kang1945.23
Kanter David2000.34
Kumar Naveen2100.34
Liao Jeffery2200.34
Deepak Narayanan23477.42
Tayo Oguntebi2436013.47
Pekhimenko Gennady2500.34
Pentecost Lillian2600.34
Vijay Janapa Reddi272931140.26
Robie Taylor2800.34
John Tom St.2900.34
Wu Carole-Jean3000.34
Xu Lingjie3100.34
Matei Zaharia329101407.89
Zaharia Matei3300.34