Explore an alternative approach to proving calculations correct in this 52-minute conference talk by James Davenport from the University of Bath. Delve into the challenges of proving algorithm and implementation correctness, and discover a method where the implementation produces intermediate information during calculations. Learn how this information can be used by proof engines to construct proofs of correctness for specific calculations, rather than general algorithms. Examine this methodology in the context of real algebraic geometry, also known as NRA in SMT-language. Gain insights into advanced mathematical concepts and their applications in machine-assisted proofs, presented at IPAM's Machine Assisted Proofs Workshop at UCLA.
How to Prove a Calculation Correct? - IPAM at UCLA
Institute for Pure & Applied Mathematics (IPAM) via YouTube
Overview
Syllabus
James Davenport - How to prove a calculation correct? - IPAM at UCLA
Taught by
Institute for Pure & Applied Mathematics (IPAM)