Leo Zhou

Quantum Algorithmic Science & Engineering (QASE) Lab @UCLA ← Burke fellow @Caltech ← PhD @Harvard ← BSc @MIT | Now also on Bsky

[Leo Zhou @leologist]

9/9
My personal bet for a convincing quantum advantage in spin glasses?
Use nonlocal quantum decoders to go beyond the dynamical transition in a low-connectivity regime, where classical alternatives may be much harder to find.

Lots of exciting questions ahead!

[Leo Zhou @leologist]

Excited to share two new papers (2611.10283, 2609.40345) on quantum algorithms for spin glasses!

We show that a quantum algorithm exactly reaches a hardness threshold conjectured since the 1980s.
However, our other paper shows that this threshold can actually be surpassed. (1/9)

[Leo Zhou @leologist]

8/9
Together, these results suggest a new perspective on average-case hardness and spin glasses:
• Local quantum decoding exactly reaches the dynamical threshold.
• Nonlocal decoding can go beyond it.
• Crossing this threshold alone does not establish quantum advantage.

[Leo Zhou @leologist]

Separately, my own lab at UCLA also has postdoc positions available (starting 2027) for broader quantum computing research.

Topics: quantum algorithms, complexity, high-dimensional statistics, and fault-tolerant quantum systems.

Apply by Nov 15 via AJO 👉https://t.co/0DszNL7pJU

[Leo Zhou @leologist]

As part of the newly established NSF Institute for Fault-Tolerant Quantum Systems, Architectures and Applications (FTQSAA), we are looking for a postdoc at UCLA at the intersection of quantum computing and AI. If you are interested, please apply via AJO: https://t.co/WmUwkvor1R

[Hrant Gharibyan ⚛️ @qhrant]

🚀 I am excited to share @BlueQubitIO first heuristic quantum advantage claim -- run on @QuantinuumQC H2 and backed by a careful comparison against state-of-the-art classical simulators. 🙌

🔷What’s new?
• We introduce a protocol for designing “peaked circuits” that can encrypt a message inside a quantum circuit with 50+ qubits. 🔐
• On H2 quantum processor, our largest instance (all-to-all connectivity, ~2000 two-qubit gates) finds the target peaked bitstring in under 2 hours. ⚡️
• Best-known classical approaches—tensor networks with belief propagation, Pauli Path simulation, and others—extrapolate to years even on exascale systems (Frontier, Summit). 🖥️⏳
• Estimated ~10,000× gap in energy and time (quantum ≈ hours vs. classical ≈ years). ♻️⚙️

🔷Why it matters
• Toward verifiable quantum advantage: a practical, testable path—plus extensive numerics supporting our extrapolations. ✅
• Benchmarking tool for NISQ hardware 🧪
• Crypto angle: potentially new type of encryption method🔒

🔷Open to the community
We’re releasing the circuits so others can try fresh classical techniques and pressure-test the gap. If you beat our baselines—or even narrow them—we want to hear from you! 🤝

📄 Check out the full manuscript: https://t.co/5a4bV6sCSC

🙏 Huge thanks to the @BlueQubitIO team for amazing work over the last couple of months, and to the @QuantinuumQC team for many insightful discussions and access to their very powerful quantum system.

#quantumcomputing #quantumadvantage #Quantinuum #NISQ #cryptography #HPC #PauliPath #tensornetworks #BlueQubit

[Leo Zhou @leologist]

The Quantum Algorithmic Science & Engineering (QASE) Lab at UCLA is hiring a postdoc (start 2026)!

Topics: quantum algorithms, complexity, high-dimensional statistics, fault-tolerant q. systems and applications

Apply by Dec 1 via AJO 👉 https://t.co/zF4gqfuI74

Please share!

[Leo Zhou @leologist]

Our team, Los Quanta, made up of @UCLA researchers, is a Semifinalist in the @XPRIZE #QuantumApplications competition!

133 teams from 31 countries are competing for a $5M prize, and we're thrilled be among the 20 advancing to the next stage!

https://t.co/zSIT5TTXfP

[John Preskill @preskill]

Aside from Yang-Mills theory and parity nonconservation, Yang made so many profound contributions to physics! One that deeply impressed me as I was starting grad school: a 1975 paper with T. T. Wu highlighting the role of fiber bundles in gauge theory.
https://t.co/kKP738L2sy

[Leo Zhou @leologist]

Perk # 15 of @UCLA : you get to hear Terry Tao’s inspirational and informative public lecture on “Machine Assisted Mathematics”

[Hrant Gharibyan ⚛️ @qhrant]

🎉 Big news! BlueQubit has closed a $10M Seed round led by Nyca Partners, with support from Chaac Ventures, Restive Ventures, & others. 🚀

We're building tools & algorithms to unlock quantum advantage.

Huge thanks to our team, customers, & investors!

PS: We're hiring! 💙 #QuantumComputing #QuantumTech #QuantumAdvantage

https://t.co/tcFnmqnpr7

[Jiao Sun @sunjiao123sun_]

Mitigating racial bias from LLMs is a lot easier than removing it from humans!

Can’t believe this happened at the best AI conference @NeurIPSConf

We have ethical reviews for authors, but missed it for invited speakers? 😡

[Leo Zhou @leologist]

Congratulations to my co-authors on finishing this nice work proposing HamQAOA, a quantum algo to optimize quantum Hamiltonians with provable guarantees! Ishaan did most of the writing here while juggling a full-time job and grad school applications. Expect great things from him!

[Ishaan Kannan @Ishaan_Kannan]

Preparing ground states of multi-particle quantum systems with local interactions is really hard.

In work (https://t.co/evsKqhz72U) with @robbieking1000 and @leologist, we propose a new, surprisingly effective quantum approximation algorithm for this QMA-hard challenge🤖⚡️.

[Leo Zhou @leologist]

Just landed in the beautiful Vancouver to attend #NeurIPS for the first time! Here to present my work on a quantum algorithm for learning tensor data with Joao Basso and @Song__Mei. Come say hi at our Spotlight poster Wednesday 11a-2p!