Towards GPU-accelerated automated reasoning- RuleML webinar (this Friday)

This is a talk/webinar announcement. The upcoming RuleML webinar will host a speaker on an ILP topic which might be of interest to some on the ilp-news mailing list.

Title: "Towards GPU-accelerated automated reasoning"

Date and time: November 15  (Friday), starting at 10 am CET

Speaker: Martin Berger<https://martinfriedrichberger.net/> (University of Sussex, UK & Montanarius Ltd).

Abstract: Graphics Processing Units (GPUs) are the work-horses of high-performance computing. The acceleration they provide to applications compatible with their programming paradigm can surpass CPU performance by several orders of magnitude, as notably evidenced by the advancements in deep learning. A significant spectrum of applications, especially within automated reasoning—like SAT/SMT solvers—has yet to reap the benefits of GPU acceleration.  In this talk, we discuss recent work that successfully implemented program synthesis on GPUs and used it to accelerate learning of logical specifications from examples.  We conclude by mapping out a research programme to move more formal verification workloads to GPUs.

Speaker's bio: Martin Berger did his PhD in formal models for distributed systems at Imperial College. He's currently an associate professor in the Department of Informatics at the University of Sussex.  He's also working as a verification consultant for the microprocessor industry and is one of the maintainers of the official RISC-V instruction set architecture (https://github.com/riscv/sail-riscv).  His research interests include logic and verification, typing systems, process calculus, meta-programming, and JIT compilers.

Webinar connection information: ZOOM
https://bit.ly/kegseminar20241115

The talk is hybrid. It will also be given in person as part of the Knowledge Engineering Group<https://kizi.vse.cz/english/seminars-keg/> seminar at VSE Prague.

-----------------
Tomas Kliegr
Organizer of the RuleML webinar<https://github.com/RuleML/ruleml-website/blob/master/talks/README.md>