Workshop on Compute Platforms for Autonomous Vehicles
To be held along with 29th ACM International Conference on Architectural Support for Programming Language and Operating Systems (ASPLOS'24)
April 27, 2024
Program
8:30 AM - 8:35 AM: Introduction
8:35 AM - 9:20 AM: Algorithm and Hardware Co-Design for Efficient Robotic Motion Planning - Bo Yuan (Rutgers) <slides>
9:20 AM - 9:40 AM: MobiTT: Speed-Oriented DNN Compression for Mobile with Tensor Train Decomposition - Miao Yin (UT Arlington), Wei Niu (U Georgia) <paper> <slides>
9:40 AM - 10:00 AM: VeCBench: A Benchmark Suite for Vehicle Computing - Tianze Wu (Institute of Computing Technology, CAS), Yongtao Yao, Qiren Wang, Arpan Bhattacharjee (U Delaware), Sa Wang, Yungang Bao (Institute of Computing Technology, CAS), Weisong Shi (U Delaware) <paper> <slides>
10:00 AM - 10:30 AM: Break
10:30 AM - 11:15 AM: Automotive Benchmarks for MLCommons - Rod Shojaei (UC Davis) <slides>
11:15 AM - 12:00 PM: Multi-Threading Compute Systems for Automotive Markets - IP Overview - Pradeep Bardia (MIPS) <slides>
About the Workshop
The global market for autonomous vehicles is expected to surpass $300 billion by the end of the decade. Due to the unique requirements of simultaneously providing sufficient computational power to run the necessary workloads needed to support autonomous driving, achieving the efficiency necessary to run the system in the resource-constrained environment of an on-road vehicle, and ensuring reliable execution, specialized compute platforms are required for this domain.
We welcome submissions focused on compute platforms for autonomous vehicles (cars/drones/robots). Key topics that we seek to address in the workshop include:
Energy-efficient machine learning inference accelerators
Techniques for exploiting reduced precision or sparsity in hardware
Techniques for improving energy efficiency of hardware
Techniques for maintaining operation/performance in an expanded thermal range
Techniques to ensure deterministic performance
Reliability and safety considerations
Call for Papers
We solicit both full papers (8-10 pages), as well as short/position papers (2-4 pages). Submissions are not double-blind (author names must be included). The page limit includes figures, tables, and appendices, but excludes references. Please use standard IEEE Word or LaTex templates. All submissions will need to be made via EasyChair.
Each submission will be reviewed by at least three reviewers from the program committee. Papers will be reviewed for novelty, quality, technical strength, and relevance to the workshop. All accepted papers will be made available online.
Please direct any questions to tomstjohn617@gmail.com
Submission Deadline: 3/22/2024
Acceptance Notification: 4/1/2024
Camera-Ready Version Due: 4/12/2024
Program Committee
Mohammad Bakhshalipour (Carnegie Mellon University)
Mostafa El-Khamy (Samsung)
William He (University of Delaware)
Liangkai Liu (University of Michigan)
Biswadip Maity (Zoox)
Kasper Mecklenburg (Arm)
Armin Runge (Ford Motor Company)
Radoyeh Shojaei (UC Davis)
Matthew Stewart (Harvard)
Arun Tejusve Raghunath Rajan (Cruise)
Zishen Wan (Georgia Tech)
Prior Events
Organizers
Tom St. John
Tom St. John is a software engineer at Meta AI where he serves as training performance technical lead for the MTIA software team. He also serves as the chair of the MLCommons automotive advisory board. Prior to his current role, he served as a technical lead for Compute Platforms at Cruise and led the distributed machine learning performance optimization efforts within Tesla Autopilot. His research primarily focuses on the intersection of parallel programming models and computer architecture design, and the impact that this has on high-performance and efficient machine learning.
Vijay Janapa Reddi
Vijay Janapa Reddi is an Associate Professor in John A. Paulson School of Engineering and Applied Sciences at Harvard University. Prior to joining Harvard University, he was a an Associate Professor at The University of Texas at Austin where he continues to be an Adjunct Professor. He leads the MLPerf inference benchmarking effort. His research interests include computer architecture, compilers and runtime systems, specifically in the context of mobile and edge computing systems to improve their performance, power efficiency, and reliability.
Weisong Shi
Weisong Shi is a professor and chair of the Department of Computer and Information Sciences at the University of Delaware, where he leads the Connected and Autonomous Research (CAR) Laboratory. Dr. Shi is also the Center Director of a recently funded NSF eCAT NSF IUCRC Center, focusing on Electric, Connected, and Autonomous Technology for Mobility. He is a fellow of IEEE, a distinguished scientist of ACM, and a member of the NSF CISE advisory committee.