94,178 Hits in 6.3 sec

An Artificial Neural Network Functionalized by Evolution [article]

Fabien Furfaro and Avner Bar-Hen and Geoffroy Berthelot
2022 arXiv   pre-print
The topology of artificial neural networks has a significant effect on their performance. Characterizing efficient topology is a field of promising research in Artificial Intelligence.  ...  However, it is not a trivial task and it is mainly experimented on through convolutional neural networks.  ...  In our case, the artificial neural network is a feed-forward network, computed from a graph [48] .  ... 
arXiv:2205.10118v1 fatcat:5sjntwmllbemjcfygpfec4gsue

Neuroevolutionary optimization [article]

Eva Volna
2010 arXiv   pre-print
This paper presents an application of evolutionary search procedures to artificial neural networks.  ...  Here, we can distinguish among three kinds of evolution in artificial neural networks, i.e. the evolution of connection weights, of architectures, and of learning rules.  ...  In principle, transfer functions of different neurons in an artificial neural network can be different and decided automatically by an evolutionary process, instead of assigned by human experts.  ... 
arXiv:1004.3557v1 fatcat:o6ilsftwlvhyhn6cw67o44tspu

Training of artificial neural networks using differential evolution algorithm

Adam Slowik, Michal Bialko
2008 2008 Conference on Human System Interactions  
In the paper an application of differential evolution algorithm to training of artificial neural networks is presented.  ...  Keywords -artificial intelligence, artificial neural network, differential evolution algorithm, training method.  ...  INTRODUCTION O apply an artificial neural network to any problem solving, it is first necessary to train the network.  ... 
doi:10.1109/hsi.2008.4581409 fatcat:gfs4asdr7nefloeclvilrcevdu

Direct fit: an ecological perspective on biological and artificial neural networks

Samuel A. Nastase
2020 Zenodo  
In this talk, we draw evolutionary theory and ecological psychology as a lens for understanding biological and artificial neural networks.  ...  Modern artificial neural networks have revealed the power in optimizing millions of parameters over millions of observations to solve real-world tasks.  ...  Direct fit to nature: an evolutionary perspective on biological and artificial neural networks. Neuron, 105 (3) , 416-434.  ... 
doi:10.5281/zenodo.4016403 fatcat:cqsojhhfibhpbip57c2f7udqt4

Dynamic Changes of Population Size in Trai ning of Artificial Neural Networks [chapter]

A. Słowik, M. Białko
2009 Human-Computer Systems Interaction  
The proposed algorithm is used to train artificial neural networks.  ...  In this paper an adaptive differential evolution algorithm with dynamic changes of population size is presented.  ...  InFigure 1aa part of an artificial neural network with neurons from n to m is shown.  ... 
doi:10.1007/978-3-642-03202-8_41 fatcat:4hm5ztrlibfjxfxftvhxv4xzie

Classification by Ensembles of Neural Networks [article]

S.V. Kozyrev
2012 arXiv   pre-print
We introduce a new procedure for training of artificial neural networks by using the approximation of an objective function by arithmetic mean of an ensemble of selected randomly generated neural networks  ...  In particular, any neural network from the mentioned ensemble may not be an approximation of the objective function.  ...  Acknowledgments The author gratefully acknowledges being partially supported by the grants of the Russian Foundation for Basic Research RFBR 11-01-00828-a and 11-01-12114-ofi-  ... 
arXiv:1202.4170v1 fatcat:hzkwvx7p4ffxdhhmn73wux2tfi

Combining Multiple Inputs in HyperNEAT Mobile Agent Controller [chapter]

Jan Drchal, Ondrej Kapral, Jan Koutník, Miroslav Šnorek
2009 Lecture Notes in Computer Science  
In this paper we present neuro-evolution of neural network controllers for mobile agents in a simulated environment.  ...  The controller is obtained through evolution of hypercube encoded weights of recurrent neural networks (HyperNEAT). The simulated agent's goal is to find a target in a shortest time interval.  ...  Acknowledgement This work has been supported by the research program "Transdisciplinary Research in the Area of Biomedical Engineering II" (MSM6840770012) sponsored by the Ministry of Education, Youth  ... 
doi:10.1007/978-3-642-04277-5_78 fatcat:mmpyt7zqkrhpnnf65ak7uki7m4

Artificial neural networks and the study of evolution of prey coloration

S. Merilaita
2007 Philosophical Transactions of the Royal Society of London. Biological Sciences  
In this paper, I investigate the use of artificial neural networks in the study of prey coloration.  ...  I conclude that visual information processing by predators is central in evolution of prey coloration.  ...  This study was supported by the Swedish Research Council.  ... 
doi:10.1098/rstb.2006.1969 pmid:17255017 pmcid:PMC2323560 fatcat:qaqa65f2zvd6fg65pmi3mztsje

AMSOM: artificial metaplasticity in SOM neural networks—application to MIT-BIH arrhythmias database

Santiago Torres-Alegre, Juan Fombellida, Juan Antonio Piñuela-Izquierdo, Diego Andina
2018 Neural computing & applications (Print)  
Now, for the first time, this kind of artificial metaplasticity is implemented in an unsupervised neural network, achieving also excellent results that are presented in this paper.  ...  Artificial metaplasticity is the machine learning algorithm inspired in the biological metaplasticity of neural synapses.  ...  A proposal of artificial metaplasticity (AMP) was generally introduced by Andina et al. [4] and detailed for supervised artificial neural networks (ANN).  ... 
doi:10.1007/s00521-018-3576-0 fatcat:s7afh324d5dpnaoydiii6tapcy

Data fusion in the decision-making process based on artificial neural networks

Janusz DUDCZYK, Institute of Computer Science and Technology, Stefan Batory State University, RYBAK RYBAK, Zdzisław JEZIERSKI, Institute of Computer Science and Technology, Stefan Batory State University, Institute of Security Sciences, Stefan Batory State University
2020 Scientific Papers of Silesian University of Technology. Organization and Management Series  
Design/methodology/approach: The conducted experiment was concerned with modelling artificial neural network to form radiation beam of microstrip antenna.  ...  The purpose of this article was an implementation of a neural network and its adaptation in the process of data fusion and solving the value prediction problem.  ...  Figure 8 presents in general the received results for six different activation functions of an artificial neural network.  ... 
doi:10.29119/1641-3466.2020.149.10 fatcat:ckqdxoyafnd6fbzlwtslc27zou

Selection Pressure and an Efficiency of Neural Network Architecture Evolving [chapter]

Halina Kwaśnicka, Mariusz Paradowski
2004 Lecture Notes in Computer Science  
The success of artificial neural network evolution is determined by many factors. One of these factors is the fitness function used in genetic algorithm.  ...  Fitness function determines selection pressure and Therefore influences the direction of evolution. It decides, whether received artificial neural network will be able to fulfill its tasks.  ...  The algorithm of evolution of artificial neural network architectures is based on the algorithm schema used in GA [?] .  ... 
doi:10.1007/978-3-540-24844-6_65 fatcat:unterilvzndwnplomzuhbwglze

Efficient evolutionary optimization using individual-based evolution control and neural networks: A comparative study

Lars Gräning, Yaochu Jin, Bernhard Sendhoff
2005 The European Symposium on Artificial Neural Networks  
This paper compares four individual-based evolution control frameworks on three widely used test functions. Feedforward neural networks are employed for fitness estimation.  ...  Second, structure optimization of neural networks mostly improves the performance of all compared algorithms.  ...  The evaluation results are used to train the neural network before the fitness of the remaining λ'−λ * offspring is estimated by the neural network.  ... 
dblp:conf/esann/GraningJS05 fatcat:atyc6tkxsbfcfokdrfp67lifee

Prediction of Ultrasonic Parameters of Mortar by using Artificial Neural Networks Techniques

Abdelilah Dariouchy, Hicham Lotfi
2017 International Journal of Computer Applications  
Two artificial neural networks (ANN) models are developed to predict the evolution of group velocity and peak to peak amplitude of an ultrasonic wave propagate in mortar, also we these models use it to  ...  Several network configurations are evaluated. For the two architectures of models, the optimal model selected is an ANN with only one hidden layer.  ...  Once the data is normalized, it remains to determine the artificial neural network architecture.  ... 
doi:10.5120/ijca2017913547 fatcat:usaanhs6gzas5lfcjqanw45fuu

Evolution of Layer Based Neural Networks

Edward R. Pantridge, Lee Spector
2016 Proceedings of the 2016 on Genetic and Evolutionary Computation Conference Companion - GECCO '16 Companion  
Modern applications of Artificial Neural Networks (ANNs) largely feature networks organized into layers of nodes.  ...  The evolution in these systems built networks on a node-by-node basis, limiting the probability of larger, layered topologies.  ...  Acknowledgments This material is based upon work supported by the National Science Foundation under Grants No. 1129139 and 1331283.  ... 
doi:10.1145/2908961.2931664 dblp:conf/gecco/PantridgeS16 fatcat:bvbdtthwnnda3dow5cxonawybq

Born to learn: The inspiration, progress, and future of evolved plastic artificial neural networks

Andrea Soltoggio, Kenneth O. Stanley, Sebastian Risi
2018 Neural Networks  
Inspired by such intricate natural phenomena, Evolved Plastic Artificial Neural Networks (EPANNs) use simulated evolution in-silico to breed plastic neural networks with a large variety of dynamics, architectures  ...  Biological plastic neural networks are systems of extraordinary computational capabilities shaped by evolution, development, and lifetime learning.  ...  Abraham (2004) proposed a method called Meta-Learning Evolutionary Artificial Neural Networks (MLEANN) in which evolution searches for initial weights, neural architectures and transfer functions for  ... 
doi:10.1016/j.neunet.2018.07.013 pmid:30142505 fatcat:gfgtbfalrbh6hnjwwj5lxj2dla
« Previous Showing results 1 — 15 out of 94,178 results