Skip to main navigation Skip to search Skip to main content

Poster Abstract: Neural Architecture Sizing for Autonomous Systems

  • Shengjie Xu
  • , Clara Hobbs
  • , Yukai Song
  • , Bineet Ghosh
  • , Sharmin Aktar
  • , Lei Yang
  • , Yi Sheng
  • , Weiwen Jiang
  • , Jingtong Hu
  • , Parasara Sridhar Duggirala
  • , Samarjit Chakraborty
  • University of North Carolina
  • University of Pittsburgh
  • The University of Alabama
  • George Mason University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

Neural networks (NNs) are now widely used for perception processing in autonomous systems. Data from sensors like cameras and lidars, after being processed by NNs, feed control algorithms that form the core of autonomy-related functions. Such NNs are implemented on graphics processing units (GPUs) and modern GPUs can be partitioned into multiple virtual machines, each implementing a separate NN. Given an autonomous system with multiple NNs, how should each NN be sized and the GPU implementing them be optimally partitioned? In this work, we study multiple GPU partitioning techniques with the goal of optimal and safe system-level control performance.

Original languageEnglish
Title of host publicationProceedings - 15th ACM/IEEE International Conference on Cyber-Physical Systems, ICCPS 2024
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages289-290
Number of pages2
ISBN (Electronic)9798350369274
DOIs
StatePublished - 2024
Externally publishedYes
Event15th Annual ACM/IEEE International Conference on Cyber-Physical Systems, ICCPS 2024 - Hong Kong, China
Duration: 13 May 202416 May 2024

Publication series

NameProceedings - 15th ACM/IEEE International Conference on Cyber-Physical Systems, ICCPS 2024

Conference

Conference15th Annual ACM/IEEE International Conference on Cyber-Physical Systems, ICCPS 2024
Country/TerritoryChina
CityHong Kong
Period13/05/2416/05/24

Keywords

  • autonomous systems
  • learning-enabled cyber-physical systems
  • neural architecture sizing
  • reachability
  • uncertainty

Fingerprint

Dive into the research topics of 'Poster Abstract: Neural Architecture Sizing for Autonomous Systems'. Together they form a unique fingerprint.

Cite this