Conference proceeding
Reliable edge machine learning hardware for scientific applications
2024 IEEE 42nd VLSI Test Symposium (VTS), pp 1-5
22 Apr 2024
Featured in Collection : UN Sustainable Development Goals @ Drexel
Abstract
Extreme data rate scientific experiments create massive amounts of data that require efficient ML edge processing. This leads to unique validation challenges for VLSI implementations of ML algorithms: enabling bit-accurate functional simulations for performance validation in experimental software frameworks, verifying those ML models are robust under extreme quantization and pruning, and enabling ultra-fine-grained model inspection for efficient fault tolerance. We discuss approaches to developing and validating reliable algorithms at the scientific edge under such strict latency, resource, power, and area requirements in extreme experimental environments. We study metrics for developing robust algorithms, present preliminary results and mitigation strategies, and conclude with an outlook of these and future directions of research towards the longer-term goal of developing autonomous scientific experimentation methods for accelerated scientific discovery.
Metrics
Details
- Title
- Reliable edge machine learning hardware for scientific applications
- Creators
- Tommaso Baldi - Fermi National Accelerator LaboratoryJavier Campos - Fermi National Accelerator LaboratoryBen Hawks - Fermi National Accelerator LaboratoryJennifer Ngadiuba - Fermi National Accelerator LaboratoryNhan Tran - Fermi National Accelerator LaboratoryDaniel Diaz - UC San Diego Health SystemJavier Duarte - UC San Diego Health SystemRyan Kastner - UC San Diego Health SystemAndres Meza - UC San Diego Health SystemMelissa Quinnan - UC San Diego Health SystemOlivia Weng - UC San Diego Health SystemCaleb Geniesse - International Computer Science InstituteAmir Gholami - International Computer Science InstituteMichael W. Mahoney - International Computer Science InstituteVladimir Loncar - Moscow Institute of Thermal TechnologyPhilip Harris - MITJoshua Agar - Drexel UniversityShuyu Qin - Drexel University
- Publication Details
- 2024 IEEE 42nd VLSI Test Symposium (VTS), pp 1-5
- Publisher
- IEEE; NEW YORK
- Number of pages
- 6
- Grant note
- National Science Foundation (10.13039/100000001)
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Mechanical Engineering and Mechanics
- Web of Science ID
- WOS:001239933000012
- Scopus ID
- 2-s2.0-85195242309
- Other Identifier
- 991021884589104721
UN Sustainable Development Goals (SDGs)
This publication has contributed to the advancement of the following goals:
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Collaboration types
- Domestic collaboration
- International collaboration
- Web of Science research areas
- Computer Science, Hardware & Architecture
- Computer Science, Theory & Methods
- Engineering, Electrical & Electronic