Applications and Techniques for Fast Machine Learning in Science

Frontiers in Big Data

Big Data

Sougata Sen¹  Justin Selig²  Yaling Liu³  Giuseppe Di Guglielmo⁴  Babar Khan⁵  Ricardo Vilalta⁶  Amir Gholami⁷  Sehoon Kim⁷  Zhen Dong⁷  Seda Ogrenci-Memik⁸  Dongning Guo⁸  Dmitri Strukov⁹  Scott Hauck^1,10  Seyedramin Rasoulinezhad^1,11  Joshua Agar^1,12  Thomas K. Warburton^1,13  Mia Liu^1,14  Tomás E. Müller Bravo^1,15  Kate Scholberg^1,16  Savannah Thais^1,17  William Tang^1,17  Allison McCarn Deiana^1,18  Javier Duarte^1,19  Shen Wang^2,20  Kin Ho Lo^2,20  Mark S. Neubauer^2,21  Gianantonio Pezzullo^2,22  Maurizio Pierini^2,23  Ekaterina Govorkova^2,23  Thea Aarrestad^2,23  Ryan A. Rivera^2,24  Kyle J. Hazelwood^2,24  Jennifer Ngadiuba^2,24  Christian Herwig^2,24  Thomas Klijnsma^2,24  Nhan Tran^2,25  Nick Fritzsche^2,26  Anne-Sophie Berthold^2,26  Kai Lukas Unger^2,27  Jürgen Becker^2,27  Steffen Bähr^2,27  Belina von Krosigk^2,28  Richard J. Bonventre^2,29  Tri Nguyen^3,30  Philip Harris^3,30  Markus Diefenthaler^3,31  Michaela Blott^3,32 

[1] Birla Institute of Technology and Science, Pilani, India;Cerebras Systems, Sunnyvale, CA, United States;Department of Bioengineering, Lehigh University, Bethlehem, PA, United States;Department of Computer Science, Columbia University, New York, NY, United States;Department of Computer Science, Technical University Darmstadt, Darmstadt, Germany;Department of Computer Science, University of Houston, Houston, TX, United States;Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA, United States;Department of Electrical and Computer Engineering, Northwestern University, Evanston, IL, United States;Department of Electrical and Computer Engineering, University of California, Santa Barbara, Santa Barbara, CA, United States;Department of Electrical and Computer Engineering, University of Washington, Seattle, WA, United States;Department of Engineering and IT, University of Sydney, Camperdown, NSW, Australia;Department of Materials Science and Engineering, Lehigh University, Bethlehem, PA, United States;Department of Physics and Astronomy, Iowa State University, Ames, IA, United States;Department of Physics and Astronomy, Purdue University, West Lafayette, IN, United States;Department of Physics and Astronomy, University of Southampton, Southampton, United Kingdom;Department of Physics, Duke University, Durham, NC, United States;Department of Physics, Princeton University, Princeton, NJ, United States;Department of Physics, Southern Methodist University, Dallas, TX, United States;Department of Physics, University of California, San Diego, San Diego, CA, United States;Department of Physics, University of Florida, Gainesville, FL, United States;Department of Physics, University of Illinois Urbana-Champaign, Champaign, IL, United States;Department of Physics, Yale University, New Haven, CT, United States;European Organization for Nuclear Research (CERN), Meyrin, Switzerland;Fermi National Accelerator Laboratory, Batavia, IL, United States;Fermi National Accelerator Laboratory, Batavia, IL, United States;Department of Electrical and Computer Engineering, Northwestern University, Evanston, IL, United States;Institute of Nuclear and Particle Physics, Technische Universität Dresden, Dresden, Germany;Karlsruhe Institute of Technology, Karlsruhe, Germany;Karlsruhe Institute of Technology, Karlsruhe, Germany;Department of Physics, Universität Hamburg, Hamburg, Germany;Lawrence Berkeley National Laboratory, Berkeley, CA, United States;Massachusetts Institute of Technology, Cambridge, MA, United States;Thomas Jefferson National Accelerator Facility, Newport News, VA, United States;Xilinx Research, Dublin, Ireland;

关键词: machine learning for science;    big data;    particle physics;    codesign;    coprocessors;    heterogeneous computing;    fast machine learning;   

DOI  :  10.3389/fdata.2022.787421

 received in 2021-09-30, accepted in 2022-01-31,  发布年份 2022

来源: Frontiers

【 摘 要 】

In this community review report, we discuss applications and techniques for fast machine learning (ML) in science—the concept of integrating powerful ML methods into the real-time experimental data processing loop to accelerate scientific discovery. The material for the report builds on two workshops held by the Fast ML for Science community and covers three main areas: applications for fast ML across a number of scientific domains; techniques for training and implementing performant and resource-efficient ML algorithms; and computing architectures, platforms, and technologies for deploying these algorithms. We also present overlapping challenges across the multiple scientific domains where common solutions can be found. This community report is intended to give plenty of examples and inspiration for scientific discovery through integrated and accelerated ML solutions. This is followed by a high-level overview and organization of technical advances, including an abundance of pointers to source material, which can enable these breakthroughs.

【 授权许可】

Unknown   
Copyright © 2022 Deiana, Tran, Agar, Blott, Di Guglielmo, Duarte, Harris, Hauck, Liu, Neubauer, Ngadiuba, Ogrenci-Memik, Pierini, Aarrestad, Bähr, Becker, Berthold, Bonventre, Müller Bravo, Diefenthaler, Dong, Fritzsche, Gholami, Govorkova, Guo, Hazelwood, Herwig, Khan, Kim, Klijnsma, Liu, Lo, Nguyen, Pezzullo, Rasoulinezhad, Rivera, Scholberg, Selig, Sen, Strukov, Tang, Thais, Unger, Vilalta, von Krosigk, Wang and Warburton.