Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Quantum Speedups for Minimax Optimization and Beyond
Authors: Chengchang Liu, Zongqi Wan, Institute of Computing Jialin Zhang, Institute of Computing Xiaoming Sun, John C. S. Lui
NeurIPS 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | This paper investigates convex-concave minimax optimization problems where only the function value access is allowed. We introduce a class of Hessian-aware quantum zeroth-order methods that can find the ǫ-saddle point within O(d2/3ǫ 2/3) function value oracle calls. This represents an improvement of d1/3ǫ 1/3 over the O(dǫ 1) upper bound of classical zeroth-order methods, where d denotes the problem dimension. We extend these results to µ-stronglyconvex µ-strongly-concave minimax problems using a restart strategy, and show a speedup of d1/3µ 1/3 compared to classical zeroth-order methods. The acceleration achieved by our methods stems from the construction of efficient quantum estimators for the Hessian and the subsequent design of efficient Hessian-aware algorithms. In addition, we apply such ideas to non-convex optimization, leading to a reduction in the query complexity compared to classical methods. |
| Researcher Affiliation | Academia | Chengchang Liu The Chinese University of Hong Kong EMAIL Zongqi Wan Great Bay University EMAIL Jialin Zhang Institute of Computing Technology, CAS EMAIL Xiaoming Sun Institute of Computing Technology, CAS EMAIL John C.S. Lui The Chinese University of Hong Kong EMAIL |
| Pseudocode | Yes | Algorithm 1 Quantum Hessian Vector(f, ǫhv, L0, L1, L2, z, v, δ) Algorithm 2 Quantum Hessian(f, ǫH, L0, L1, L2, z, δ) Algorithm 3 HAQZO(z0, T, L0, L1, L2, δ) Algorithm 4 HAQZO+(z0, T, L0, L1, L2, M, m, δ) Algorithm 5 Restart-HAQZO+(z0, T, L0, L1, L2, M, m, S, δ) Algorithm 6 QCNM(z0, T, L0, L1, L2, M, m, ǫg, ǫH, δ) |
| Open Source Code | No | The paper does not contain any explicit statements about open-sourcing code, links to repositories, or code provided in supplementary materials. |
| Open Datasets | No | The paper is theoretical and does not involve experiments or the use of specific datasets. The NeurIPS Paper Checklist also indicates 'NA' for questions related to experimental results, implicitly confirming no datasets are used. |
| Dataset Splits | No | The paper is theoretical and does not involve experiments with datasets, therefore, no dataset splits are discussed or provided. |
| Hardware Specification | No | The paper is theoretical and does not involve running experiments on hardware. Therefore, no hardware specifications are provided. |
| Software Dependencies | No | The paper is theoretical and does not involve running experiments with specific software dependencies or versions. |
| Experiment Setup | No | The paper is theoretical and focuses on algorithm design and proofs, not empirical experiments. Therefore, no experimental setup details like hyperparameters or training configurations are provided. |