Abstract:In two pervious papers \cite{dndiep3}, \cite{dndiep4}, the first author constructed the least square quantum neural networks (LS-QNN), and ploynomial interpolation quantum neural networks ( PI-QNN), parametrico-stattistical QNN like: leanr regrassion quantum neural networks (LR-QNN), polynomial regression quantum neural networks (PR-QNN), chi-squared quantum neural netowrks ($\chi^2$-QNN). We observed that the method works also in the cases by using nonparametric statistics. In this paper we analyze and implement the nonparametric tests on QNN such as: linear nonparametric regression quantum neural networks (LNR-QNN), polynomial nonparametric regression quantum neural networks (PNR-QNN). The implementation is constructed through the Gauss-Jordan Elimination quantum neural networks (GJE-QNN).The training rule is to use the high probability confidence regions or intervals.

What problem does this paper attempt to address?

### What problems does this paper attempt to solve? This paper aims to explore and implement quantum neural networks (QNN) based on non - parametric statistics. Specifically, the authors attempt to apply non - parametric regression methods to quantum neural networks to solve regression problems in machine learning. Unlike traditional parametric methods, non - parametric methods do not need to assume the specific form of data distribution, so they have greater flexibility and adaptability. #### Main problems include: 1. **Non - parametric regression**: - The authors analyze and implement linear non - parametric regression quantum neural networks (LNR - QNN) and polynomial non - parametric regression quantum neural networks (PNR - QNN). - In these models, the authors use high - probability intervals to formulate training rules instead of relying on high - probability regions of standardized test statistics. 2. **Application of quantum computing**: - The paper solves linear and polynomial regression problems through the implementation of quantum Gauss - Jordan elimination (GJE) in quantum neural networks. - The authors use quantum algorithms such as Grover's search algorithm and quantum Fourier transform (QFT) to improve computational efficiency. 3. **Training rules**: - The authors propose training rules based on high - probability intervals to ensure that the model's prediction results fall within a given confidence interval. This rule is represented by the following formula: \[ P(a(x)\leq r(x)\leq b(x)) = 1-\alpha \] - Here, $a(x)$ and $b(x)$ are the upper and lower bounds of the confidence interval, and $\alpha$ is the significance level. 4. **Implementation methods**: - The authors describe in detail how to implement non - parametric regression in quantum neural networks, including linear non - parametric regression and polynomial non - parametric regression. - Use quantum Gauss - Jordan elimination (QGJE) to solve linear equations and find pivot columns through Grover's search algorithm. ### Summary The main contribution of this paper is to introduce non - parametric statistical methods into quantum neural networks, propose a new training rule, and improve the efficiency and accuracy of the model through quantum computing techniques. This provides a new idea and tool for the field of quantum machine learning, especially when dealing with complex data distributions.

Nonparametric Regression Quantum Neural Networks

Statistical Tests and Confidential Intervals as Thresholds for Quantum Neural Networks

Quantum Similarity Testing with Convolutional Neural Networks.

Predictive Modelling of Quantum Process with Neural Networks

A Quantum-Implementable Neural Network Model

Autoregressive neural quantum states of Fermi Hubbard models

Statistical Analysis of Quantum State Learning Process in Quantum Neural Networks

Quantum Algorithms for Feedforward Neural Networks

Non-parametric Greedy Optimization of Parametric Quantum Circuits

Building Quantum Neural Networks Based on a Swap Test

Quantum Generative Training Using Rényi Divergences

The Dilemma of Quantum Neural Networks

Reduction of finite sampling noise in quantum neural networks

Quantum Neural Network for Quantum Neural Computing

Quantum Neural Network Classifiers: A Tutorial

Theoretical error performance analysis for variational quantum circuit based functional regression

Training Classical Neural Networks by Quantum Machine Learning

Coherent Feed Forward Quantum Neural Network

Analyzing Convergence in Quantum Neural Networks: Deviations from Neural Tangent Kernels

Deep quantum neural networks form Gaussian processes