Q2B25 Silicon Valley | Edward Farhi, Principal Scientist, Google

Item: Q2B25 Silicon Valley | Edward Farhi, Principal Scientist, Google
Rating: 4
Author: PostSingularity Institute

by Edward Farhi

Formal & Physical Sciences Computing & Cybersecurity UYComputer science UYXNatural computing

🎙 Edward Farhi 👥 6K 📅 January 7, 2026 ⏱ 19 min 👁 173 📄 expert opinion 🧭 2026-08-16

Available in: English (current) Français

Keywords

QAOAquantum approximate optimizationcombinatorial optimizationquantum supremacyspin glass

Summary

Edward Farhi, Principal Scientist at Google and co-inventor of the Quantum Approximate Optimization Algorithm (QAOA), presents the algorithm’s key features and recent results. He highlights its worst-case performance guarantees, quantum supremacy at shallow depth, landscape independence, and effectiveness on typical instances. He explains the algorithm’s structure: alternating unitaries parametrized by gamma and beta, applied to a uniform superposition. He discusses the Max-3-Lin-2 problem, where QAOA achieves an exponential speedup over classical algorithms, and the Sherrington-Kirkpatrick (SK) model, where QAOA’s performance can be analyzed in the infinite-size limit. He shows that QAOA parameters fall on universal curves, enabling efficient parameter setting. He concludes that QAOA improves with depth, with no known barriers, and suggests it may outperform classical algorithms on near-term quantum computers.

122 words

Critical Evaluation

Value of the Information & Strength of the Argument

The talk provides valuable insights into QAOA’s theoretical guarantees and practical performance. Farhi presents provable results, such as worst-case guarantees and quantum supremacy at depth one, which are rare in quantum optimization. He also shows numerical evidence of QAOA outperforming other algorithms on specific problems. The argumentation is solid, relying on published results and comparisons with classical algorithms. However, some claims, such as the universal parameter curves, are presented as observations without full theoretical explanation, and the talk is more of an expert overview than a rigorous proof.

97 words

Title / Content Match

The title accurately reflects the content: a presentation by Edward Farhi on the QAOA at the Q2B conference.

Quality & Reliability

8/10

Talk by a leading researcher (Edward Farhi, inventor of QAOA) presenting results with provable guarantees and comparisons to classical algorithms. Some claims are anecdotal or based on unpublished work, but the core content is scientifically grounded.

Key Moments

Markers derived by PSI from the transcript: the creator did not define chapters.

Cited Sources

Q2B Conference Website — Conference where the talk was given
Slides for the talk — Presentation slides referenced in the video description

Concurring Sources

Quantum Approximate Optimization Algorithm — Original paper introducing QAOA

Contribution & Novelties

The talk provides an expert perspective on QAOA, highlighting recent results on provable performance guarantees and high-depth analysis. It introduces the concept of universal parameter curves, which could simplify parameter optimization. The talk also emphasizes QAOA’s quantum supremacy at shallow depth and its potential for near-term quantum advantage.

Pour aller plus loin :

Quantum Approximate Optimization Algorithm — Original paper by Farhi, Goldstone, and Gutmann.
Sherrington-Kirkpatrick model — Spin glass model discussed in the talk.
Quantum supremacy — Concept of quantum advantage over classical computers.

85 words

Radar Profile

The radar profile shows high scores across all dimensions, indicating a technically dense and reliable presentation. The talk is particularly strong in information quantity and technical level, with a slightly lower but still high score in global reliability due to some anecdotal claims.

Reliability 8/10