English

Regularization of linear machine learning problems

Numerical Analysis 2024-08-12 v1 Numerical Analysis

Abstract

In this paper, we consider the simplest version of a linear neural network (LNN). Assuming that for training (constructing an optimal weight matrix QQ) we have a set of training pairs, i.e. we know the input data \begin{equation} G=\left\{g^{\left(1\right)},g^{\left(2\right)},\cdots,g^{\left(K\right)}\right\}, \end{equation} as well as the correct answers to these input data \begin{equation} H=\left\{h^{\left(1\right)},h^{\left(2\right)},\cdots,h^{\left(K\right)}\right\}. \end{equation} We will study the possibilities of constructing a weight matrix QQ of a neural network that will give correct answers to arbitrary input data based on the connection of the specified problem with a system of linear algebraic equations (SLAE). Consider a class of neural networks in which each neuron has only one output signal and performs linear operations. We will show how such LNEs are reduced to SLAEs. Since the questions GG and the correct answers HH are known to us, the desired weight matrix QQ must satisfy the equations \begin{equation} Qg^{\left(k\right)}=h^{\left(k\right)}, k=1,2,\cdots,K. \end{equation} It is required to restore QQ. In the general case, the matrix QQ is rectangular Q=QMN={qmn}Q=Q_{MN}=\left\{q_{mn}\right\}, mm is the row number, and g(k)RNg^{\left(k\right)}\in\mathbb{R}^N, h(k)RMh^{\left(k\right)}\in\mathbb{R}^M. Let GNKG_{NK} be a matrix composed of columns g(1),g(2),,g(k)g^{\left(1\right)}, g^{\left(2\right)},\cdots,g^{\left(k\right)}, and HMKH_{MK} be a matrix composed of columns h(1),h(2),,h(k)h^{\left(1\right)},h^{\left(2\right)},\cdots,h^{\left(k\right)}. Then, with respect to QMNQ_{MN}, we obtain a matrix SLAE. \begin{equation} Q_{MN}G_{NK}=H_{MK}. \end{equation} This paper will present methods for regularizing the constructed system.

Keywords

Cite

@article{arxiv.2408.04871,
  title  = {Regularization of linear machine learning problems},
  author = {S. Liu and S. I. Kabanikhin and S. V. Strijhak},
  journal= {arXiv preprint arXiv:2408.04871},
  year   = {2024}
}

Comments

in Russian language

R2 v1 2026-06-28T18:08:21.173Z