arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2001.09396cs.LGcs.ITcs.NEeess.SPmath.ITstat.ML
收藏

具有矩阵值未知数的多层网络中的推理

Inference in Multi-Layer Networks with Matrix-Valued Unknowns

University of California, Los Angeles(加利福尼亚大学洛杉矶分校);New York University(纽约大学);The Ohio State University(俄亥俄州立大学)

Parthe Pandit, Mojtaba Sahraee-Ardakan, Sundeep Rangan, Philip Schniter, Alyson K. Fletcher

摘要:

本文研究了从输出的观测值推断随机多层神经网络的输入和隐变量的问题。每一层中的隐变量被表示为矩阵。这个问题适用于通过深度生成先验模型、多任务和混合回归以及学习某些类的双层神经网络进行信号恢复。通过扩展多层向量近似消息传递(ML-VAMP)算法,提出了一种统一的MAP和MMSE推理的近似算法。结果表明,在未知量的维数$N\times d$按$N\rightarrow\infty$增长,且$d$固定的情况下,多层矩阵VAMP(ML-Mat-VAMP)算法的性能在一定的随机大系统极限下可以准确预测.在双层神经网络学习问题中,这种缩放对应于输入特征和训练样本的数量增长到无穷大但隐藏节点的数量保持固定的情况。该分析能够精确预测学习的参数和测试误差。

英文摘要:

We consider the problem of inferring the input and hidden variables of a stochastic multi-layer neural network from an observation of the output. The hidden variables in each layer are represented as matrices. This problem applies to signal recovery via deep generative prior models, multi-task and mixed regression and learning certain classes of two-layer neural networks. A unified approximation algorithm for both MAP and MMSE inference is proposed by extending a recently-developed Multi-Layer Vector Approximate Message Passing (ML-VAMP) algorithm to handle matrix-valued unknowns. It is shown that the performance of the proposed Multi-Layer Matrix VAMP (ML-Mat-VAMP) algorithm can be exactly predicted in a certain random large-system limit, where the dimensions $N\times d$ of the unknown quantities grow as $N\rightarrow\infty$ with $d$ fixed. In the two-layer neural-network learning problem, this scaling corresponds to the case where the number of input features and training samples grow to infinity but the number of hidden nodes stays fixed. The analysis enables a precise prediction of the parameter and test error of the learning.

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。

↑