Haimeng Zhao

@haimengzhao

PhD student in Physics @ Caltech. Tsinghua '24. Quantum Information. Quantum Many-Body Physics. Machine Learning.

Pasadena, Shanghai

hmzhao.me

Joined March 2022

420 Following

822 Followers

Replies Reposts Media

Pinned
Haimeng Zhao
@haimengzhao
Apr 10
⚛ Can small quantum computers accelerate AI on massive classical data? Yes! I am absolutely thrilled to share our new work proving *honest* exponential quantum advantages in broadly applicable classical tasks. 🧵👇 Paper: arxiv.org/abs/2604.07639 Blog: quantumfrontiers.com/2026/04/09/unl…
36
Haimeng Zhao
@haimengzhao
Apr 11, 2025
Do abstract learning processes have tangible physical consequences? Yes! In this work, we rigorously show that the (in)ability to learn dramatically impacts the amount of physical resources required to perform certain thermodynamic tasks. arxiv.org/abs/2504.07341
3
Haimeng Zhao
@haimengzhao
Oct 8, 2024
Ever questioned yourself whether quantum advantage in classical machine learning tasks is even possible? In this new paper, we give affirmative evidence to this question! We provide both theoretical proofs and empirical demonstrations on trapped ions! scirate.com/arxiv/2410.030…
7
Haimeng Zhao
@haimengzhao
May 27, 2024
New paper published: we numerically study how the sample complexity of Neural Quantum State tomography depends on the mixedness of the target states. With a general trick to boost accuracy, we illustrate how different ansatze (POVM & NDO) performs under different temperature.
Quantum
@quantumjournal
May 27, 2024
Fresh in Quantum: Empirical Sample Complexity of Neural Network Mixed State Reconstruction by Haimeng Zhao, Giuseppe Carleo, and Filippo Vicentini doi.org/10.22331/q-202…
1
Haimeng Zhao
@haimengzhao
Nov 1, 2023
Nature, as a giant quantum computer, can only prepare states and evolution of bounded complexity. Can we take advantage of this prior, and learn these states and dynamics efficiently? In this new paper, we find the answer to be a mix of Yes and No! (1/8) scirate.com/arxiv/2310.198…
3