SSTQ：基于子采样随机 TurboQuant 的隐私保护向量量化
SSTQ:Privacy-Preserving Vector Quantization via Subsampled Stochastic TurboQuant
浏览论文内容
中文总结 AI 辅助
本研究针对分布式优化中隐私保护与通信成本的矛盾，提出 SSTQ 框架，实现了更优的均方误差缩放，在联邦学习任务中展现出良好效用与通信效率。
中文摘要 AI 辅助
在分布式优化中实现局部差分隐私同时保持低通信成本仍是一项挑战。现有向量量化方法如 vqSGD 采用高维几何构造，但会产生不利的维度相关方差。本研究提出子采样随机 TurboQuant（SSTQ）框架，结合超完备等范数紧框架、坐标子采样及感知隐私的一维量化。SSTQ 包含两个变体：平坦随机响应版本和度量感知拉普拉斯版本，后者更适配高码本比特宽度场景。研究表明，SSTQ 实现了最优均方误差缩放，每个客户端仅使用 ⌈log₂N⌉ + b 比特，其中 N = Θ(d) 为框架大小；还推导了感知隐私的替代码本目标，将码本相关的均方误差缩放从 O(4ᵇ) 降至 O(2ᵇ)。最后，在使用 CIFAR-10 和 Fashion-MNIST 的联邦学习任务中，对 SSTQ 与现有基线进行实证评估，证明其具备良好的效用和通信效率。
英文摘要
Achieving local differential privacy in distributed optimization while maintaining low communication cost remains challenging. Existing vector quantization methods, such as vqSGD, rely on high-dimensional geometric constructions but incur unfavorable dimension-dependent variance. In this work, we propose Subsampled Stochastic TurboQuant (SSTQ), a framework that combines a bounded Kashin representation, data-independent coordinate subsampling, and privacy-aware one-dimensional quantization. SSTQ includes two variants: (1) a Flat Randomized Response variant that is unbiased and, for a fixed codebook bit-width, frame redundancy, and dimension-independent Kashin level, achieves reconstruction MSE that scales linearly with the ambient dimension $d$, while using only $\lceil \log_2 N \rceil + b$ bits per message. Here, $N = Θ(d)$ denotes the frame size in the Kashin transform and $b$ is the codebook bit-width; and (2) a metric-aware truncated-Laplace variant that removes the exponential dependence on bit-width at the cost of a non-vanishing bias. We also derive a convex uniform-surrogate codebook objective whose worst-case codebook-dependent upper bound improves from $O(4^b)$ to $O(2^b)$. Experiments on synthetic regression, Fashion-MNIST, and CIFAR-10 compare the per-message privacy-utility and uplink-communication trade-offs of SSTQ with those of established baselines, demonstrating favorable utility and communication efficiency.
发表机构
- University of Southern California（南加州大学）
- Google Research（谷歌研究院）
- Carnegie Mellon University（卡内基梅隆大学）
机构由 AI 辅助整理，请以论文原文为准。
