Real-Time Decoding for Fault-Tolerant Quantum Computers
Institute for Pure & Applied Mathematics (IPAM) via YouTube
Overview
Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore the critical challenge of real-time decoding for fault-tolerant quantum computers in this 44-minute conference talk from IPAM's Bridging the Gap Between NISQ and FTQC Workshop. Discover how the transition beyond NISQ (Noisy Intermediate-Scale Quantum) devices requires meeting microsecond-scale QEC (Quantum Error Correction) cycle times, necessitating specialized classical hardware like FPGAs or ASICs rather than traditional CPU-based approaches. Learn about the fundamental shift in algorithmic performance when moving to highly parallel hardware architectures, where Gaussian elimination can achieve linear parallel time on FPGAs compared to cubic time on CPUs, motivating hardware-aware redesign of decoding methods. Examine recent progress in real-time decoding solutions, with particular focus on message-passing decoders and the promising Relay-BP algorithm, which enhances standard belief propagation convergence while maintaining a lightweight, highly parallel structure ideal for FPGA implementation. Understand how Relay-BP significantly outperforms alternative decoders for quantum LDPC codes and addresses the critical backlog problem arising from variable decoding latency. Gain insights into the mathematical conditions governing decoder latency distributions that enable fast average-case decoding with sufficiently light latency tails, ensuring decoding can keep pace with syndrome generation and maintain bounded computational slowdown in large-scale fault-tolerant quantum computations.
Syllabus
Michael Beverland - Real-time decoding for fault-tolerant quantum computers - IPAM at UCLA
Taught by
Institute for Pure & Applied Mathematics (IPAM)