Benchmarks: From Maze to Compass
summary
Panel argues benchmarking must start from a clear theoretical reason why a quantum device will outperform classical methods, then build the best classical solver, run small NISQ prototypical experiments and compare against quantum volume and raw gate fidelities. Notes instability of vendor-reported numbers and calls for well‑specified classical benchmark databases and published problems (example: an NMR convergence benchmark tied to a high‑accuracy classical solver). Suggests hardware targets like >99.9% two‑qubit gate fidelity for certain chemistry mappings.
AI-written summary · about
chapters (16)
- 0:00Welcome
- 0:44Quantum computing stage
- 2:32Benchmarking areas
- 3:54Meaningful benchmarking
- 6:40Bottom-up benchmarking
- 7:01Key stakeholders
- 8:47Community input
- 9:50NMR spectroscopy benchmarks
- 11:58User feedback
- 13:24Quantum volume evaluation
- 14:29Standardizing benchmarks
- 15:14Stability challenges
- 17:19Broader benchmarking adoption
- 17:59Classical benchmark databases
- 19:03Recommendations for new users
- 20:09Call to action
related talks
20mThe Value of Quantum Benchmarking
01h 09mRecent Progress in Quantum Advantage
43mFrom lab to marketplace: How do we succeed in quantum computing?
01h 13mComputational quantum chemistry: Classical gaps, quantum opportunities, open questions
32mComputational quantum chemistry: Classical gaps, quantum opportunities, open questions Part 2
28mThe Problem with Qubits
24mQuantum Computing: a Treasure Hunt, not a Gold Rush
59mUtility, Computational Cost, and Effective Quantum Volume of NISQ Experiments