arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.11277cs.ITcs.LGeess.SPmath.ITstat.ML
收藏

线性模型的Minimax最优Quantization:信息论极限和有效算法

Minimax Optimal Quantization of Linear Models: Information-Theoretic Limits and Efficient Algorithms

Stanford University(斯坦福大学);Princeton University(普林斯顿大学)

Rajarshi Saha, Mert Pilanci, Andrea J. Goldsmith

摘要:

高维模型通常具有较大的内存占用量,并且必须在训练后进行量化,然后才能部署在资源受限的边缘设备上进行推理任务。在这项工作中,我们开发了一个信息理论框架的问题量化的线性回归从训练数据$(\mathbf{X}, \mathbf{y})$,一些基本的统计关系$\mathbf{y} = \mathbf{X}\boldsymbolθ + \mathbf{v}$。作为潜在参数$\boldsymbolθ \in \mathbb{R}^d$的估计的学习模型被约束为仅使用$Bd$位可表示,其中$B \in (0, \infty)$是预先指定的预算,并且$d$是维度。我们推导出一个信息理论下界的极大极小风险在此设置下,并提出了一个匹配的上限使用随机嵌入为基础的算法,这是紧到常数因子。下限和上限一起表征实现与未量化设置相当的性能风险所需的最小阈值比特预算。我们还提出了随机Hadamard嵌入,计算效率高,是最佳的一个温和的对数因子的下限。我们的模型量化策略可以推广,我们通过将方法和上限扩展到非线性回归的双层ReLU神经网络来展示其有效性。数值模拟表明,我们提出的计划,以及它的接近下界的改善性能。

英文摘要:

High-dimensional models often have a large memory footprint and must be quantized after training before being deployed on resource-constrained edge devices for inference tasks. In this work, we develop an information-theoretic framework for the problem of quantizing a linear regressor learned from training data $(\mathbf{X}, \mathbf{y})$, for some underlying statistical relationship $\mathbf{y} = \mathbf{X}\boldsymbolθ + \mathbf{v}$. The learned model, which is an estimate of the latent parameter $\boldsymbolθ \in \mathbb{R}^d$, is constrained to be representable using only $Bd$ bits, where $B \in (0, \infty)$ is a pre-specified budget and $d$ is the dimension. We derive an information-theoretic lower bound for the minimax risk under this setting and propose a matching upper bound using randomized embedding-based algorithms which is tight up to constant factors. The lower and upper bounds together characterize the minimum threshold bit-budget required to achieve a performance risk comparable to the unquantized setting. We also propose randomized Hadamard embeddings that are computationally efficient and are optimal up to a mild logarithmic factor of the lower bound. Our model quantization strategy can be generalized and we show its efficacy by extending the method and upper-bounds to two-layer ReLU neural networks for non-linear regression. Numerical simulations show the improved performance of our proposed scheme as well as its closeness to the lower bound.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑