用SPSA优化量子自然语言推理模型，发现其方差大导致性能远低于传统方法。
SPSA Hyperparameter Tuning for Variational Quantum Natural Language Inference
- 采用SPSA仅需两次采样计算梯度，降低计算成本。
- 最优配置下测试准确率55%，仍比参数移位法低16-19个百分点。
- 适合资源受限但容忍精度损失的量子模型调优场景。
训练变分量子模型需在精确但高成本的参数移位梯度与低成本但高方差的随机近似梯度间权衡。本文在6量子比特、60参数的量子自然语言推理（QNLI）分类器上，系统评估了SPSA扰动尺度、学习率和衰减策略的影响。最佳配置（c₀=0.01, η=0.10, γ=0.10）的测试准确率为55%±11%，优于默认配置（49%±6%），但仍比参数移位+AdamW（高出16-19个百分点）和BuresQNG显著落后。经典增益和Bures预条件的SPSA表现更差，分别仅达51%和46%，因噪声被放大。结果表明，两样本SPSA梯度的高方差限制了其在60参数任务中的可靠优化能力。
原文摘要 · Abstract (English)
Training variational quantum models requires choosing between parameter-shift gradients, which are exact but cost $O(P)$ forward evaluations, and simultaneous perturbation stochastic approximation (SPSA), which uses only two samples but produces high-variance estimates that can degrade optimisation on small supervised tasks. Whether the cheap gradient is usable depends on the variance that results from different choices of the SPSA perturbation scale, learning rate, and gain-decay schedule. We varied those quantities across a broad grid on a 6-qubit, 60-parameter QNLI classifier and compared the best configurations to parameter-shift AdamW and BuresQNG. AdamW-style SPSA with $c_0=0.01$, $η=0.10$, $γ=0.10$ reached $55\% \pm 11\%$ test accuracy, improving over the default configuration ($49\% \pm 6\%$) but remaining 16-19 percentage points below the parameter-shift baselines because the two-sample SPSA gradient estimate has too much variance for reliable optimisation of 60 parameters in 40 epochs. Classical-gain SPSA and Bures-preconditioned SPSA performed worse, at $51\%$ and $46\%$ respectively. Bures-preconditioning a noisy two-sample SPSA gradient amplifies perturbation noise.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。