학술논문
Applications of Deep Learning to physics workflows
Document Type
Working Paper
Author
Agarwal, Manan; Alameda, Jay; Audenaert, Jeroen; Benoit, Will; Beveridge, Damon; Bhattacharya, Meghna; Chatterjee, Chayan; Chatterjee, Deep; Chen, Andy; Cholayil, Muhammed Saleem; Chou, Chia-Jui; Choudhary, Sunil; Coughlin, Michael; Dax, Maximilian; Desai, Aman; Di Luca, Andrea; Duarte, Javier Mauricio; Farrell, Steven; Feng, Yongbin; Goodarzi, Pooyan; Govorkova, Ekaterina; Graham, Matthew; Guiang, Jonathan; Gunny, Alec; Guo, Weichangfeng; Hakenmueller, Janina; Hawks, Ben; Hsu, Shih-Chieh; Jawahar, Pratik; Ju, Xiangyang; Katsavounidis, Erik; Kellis, Manolis; Khoda, Elham E; Lahbabi, Fatima Zahra; Lian, Van Tha Bik; Liu, Mia; Malanchev, Konstantin; Marx, Ethan; McCormack, William Patrick; McLeod, Alistair; Mo, Geoffrey; Moreno, Eric Anton; Muthukrishna, Daniel; Narayan, Gautham; Naylor, Andrew; Neubauer, Mark; Norman, Michael; Omer, Rafia; Pedro, Kevin; Peterson, Joshua; Pürrer, Michael; Raikman, Ryan; Raj, Shivam; Ricker, George; Robbins, Jared; Samani, Batool Safarzadeh; Scholberg, Kate; Schuy, Alex; Skliris, Vasileios; Soni, Siddharth; Sravan, Niharika; Sutton, Patrick; Villar, Victoria Ashley; Wang, Xiwei; Wen, Linqing; Wuerthwein, Frank; Yang, Tingjun; Yeh, Shu-Wei
Source
Subject
Language
Abstract
Modern large-scale physics experiments create datasets with sizes and streaming rates that can exceed those from industry leaders such as Google Cloud and Netflix. Fully processing these datasets requires both sufficient compute power and efficient workflows. Recent advances in Machine Learning (ML) and Artificial Intelligence (AI) can either improve or replace existing domain-specific algorithms to increase workflow efficiency. Not only can these algorithms improve the physics performance of current algorithms, but they can often be executed more quickly, especially when run on coprocessors such as GPUs or FPGAs. In the winter of 2023, MIT hosted the Accelerating Physics with ML at MIT workshop, which brought together researchers from gravitational-wave physics, multi-messenger astrophysics, and particle physics to discuss and share current efforts to integrate ML tools into their workflows. The following white paper highlights examples of algorithms and computing frameworks discussed during this workshop and summarizes the expected computing needs for the immediate future of the involved fields.
Comment: Whitepaper resulting from Accelerating Physics with ML@MIT workshop in Jan/Feb 2023
Comment: Whitepaper resulting from Accelerating Physics with ML@MIT workshop in Jan/Feb 2023