Wyniki wyszukiwania - BazTech

1

Choice of the p-norm for high level classification features pruning in modern convolutional neural networks with local sensitivity analysis

Jeczmionek Ernest, Kowalski Piotr A.

International Journal of Applied Mathematics and Computer Science

|

2023

|

Vol. 33, no. 4

663--672

EN

Transfer learning has surfaced as a compelling technique in machine learning, enabling the transfer of knowledge across networks. This study evaluates the efficacy of ImageNet pretrained state-of-the-art networks, including DenseNet, ResNet, and VGG, in implementing transfer learning for prepruned models on compact datasets, such as FashionMNIST, CIFAR10, and CIFAR100. The primary objective is to reduce the number of neurons while preserving high-level features. To this end, local sensitivity analysis is employed alongside p-norms and various reduction levels. This investigation discovers that VGG16, a network rich in parameters, displays resilience to high-level feature pruning. Conversely, the ResNet architectures reveal an interesting pattern of increased volatility. These observations assist in identifying an optimal combination of the norm and the reduction level for each network architecture, thus offering valuable directions for model-specific optimization. This study marks a significant advance in understanding and implementing effective pruning strategies across diverse network architectures, paving the way for future research and applications.

2

A modified particle swarm optimization procedure for triggering fuzzy flip-flop neural networks

Kowalski Piotr A., Słoczyński Tomasz

International Journal of Applied Mathematics and Computer Science

|

2021

|

Vol. 31, no. 4

577--586

EN

The aim of the presented study is to investigate the application of an optimization algorithm based on swarm intelligence to the configuration of a fuzzy flip-flop neural network. Research on solving this problem consists of the following stages. The first one is to analyze the impact of the basic internal parameters of the neural network and the particle swarm optimization (PSO) algorithm. Subsequently, some modifications to the PSO algorithm are investigated. Approximations of trigonometric functions are then adopted as the main task to be performed by the neural network. As a result of the numerical verification of the problem, a set of rules are developed that can be helpful in constructing a fuzzy flip-flop type neural network. The obtained results of the computations significantly simplify the structure of the neural network in relation to similar conditions known from the literature.

3

Preface to special issue on recent advances in machine learning and its applications

Kowalski Piotr A., Łukasik Szymon, Kulczycki Piotr

Journal of Automation Mobile Robotics and Intelligent Systems

|

2020

|

Vol. 14, No. 2

30--31

4

Preface to special issue on recent advances in information technology III

Kowalski Piotr A., Łukasik Szymon

Journal of Automation Mobile Robotics and Intelligent Systems

|

2020

|

Vol. 14, No. 3

18--19

5

Preface to special issue on contemporary problems of computer science, physics and applied mathematics

Kowalski Piotr A., Łukasik Szymon, Kulczycki Piotr

Journal of Automation Mobile Robotics and Intelligent Systems

|

2019

|

Vol. 13, No. 3

39--40

6

Efficient astronomical data condensation using approximate nearest neighbors

Łukasik Szymon, Lalik Konrad, Sarna Piotr, Kowalski Piotr A., Charytanowicz Małgorzata, Kulczycki Piotr

International Journal of Applied Mathematics and Computer Science

|

2019

|

Vol. 29, no. 3

467--476

EN

Extracting useful information from astronomical observations represents one of the most challenging tasks of data exploration. This is largely due to the volume of the data acquired using advanced observational tools. While other challenges typical for the class of big data problems (like data variety) are also present, the size of datasets represents the most significant obstacle in visualization and subsequent analysis. This paper studies an efficient data condensation algorithm aimed at providing its compact representation. It is based on fast nearest neighbor calculation using tree structures and parallel processing. In addition to that, the possibility of using approximate identification of neighbors, to even further improve the algorithm time performance, is also evaluated. The properties of the proposed approach, both in terms of performance and condensation quality, are experimentally assessed on astronomical datasets related to the GAIA mission. It is concluded that the introduced technique might serve as a scalable method of alleviating the problem of the dataset size.