Abstract-The emergence of collective dynamics in neural networks is a mechanism of the animal and human brain for information processing. In this paper, we develop a computational technique using distributed processing elements in a complex network, which are called particles, to solve semi-supervised learning problems. Three actions govern the particles' dynamics: generation, walking, and absorption. Labeled vertices generate new particles that compete against rival particles for edge domination. Active particles randomly walk in the network until they are absorbed by either a rival vertex or an edge currently dominated by rival particles. The result from the model evolution consists of sets of edges arranged by the label dominance. Each set tends to form a connected subnetwork to represent a data class. Although the intrinsic dynamics of the model is a stochastic one, we prove there exists a deterministic version with largely reduced computational complexity; specifically, with linear growth. Furthermore, the edge domination process corresponds to an unfolding map in such way that edges "stretch" and "shrink" according to the vertexedge dynamics. Consequently, the unfolding effect summarizes the relevant relationships between vertices and the uncovered data classes. The proposed model captures important details of connectivity patterns over the vertex-edge dynamics evolution, in contrast to previous approaches which focused on only vertex or only edge dynamics. Computer simulations reveal that the new model can identify nonlinear features in both real and artificial data, including boundaries between distinct classes and overlapping structures of data.
This paper presents a model for a dynamical system where particles dominate edges in a complex network. The proposed dynamical system is then extended to an application on the problem of community detection and data clustering. In the case of the data clustering problem, 6 different techniques were simulated on 10 different datasets in order to compare with the proposed technique. The results show that the proposed algorithm performs well when prior knowledge of the number of clusters is known to the algorithm.
I convey my heartfelt gratitude to all the people who have helped and inspired me during my graduate studies at ICMC-USP. I would like to thank my professor, Dr. Zhao Liang, for his insights, advice, patience, dedication, and for keeping me on the line while extending, at the same time, freedom to drive decisions. To my colleague and friend, Filipe Verri, with whom significant work has been done, I am also grateful. I like to believe that, while working side by side, we have inaugurated a new form of communication marked by endless brainstorming, treating each other as sounding boards for ideas. The members of the Bio-inspired Computing Laboratory have been an invaluable resource for ideas and assistance, in particular, Daniel Cestari, for both his insights and patience. Though far away, my family is my source of strength, and has always supported and advised me over the years toward a meaningful life. Mom, Dad, Angela, Ricardo, Vlade, Victor, Laura, and Quim, thank you. Finally, I thank the university staff for being so helpful and am grateful for the financial support offered by the Coordination for the Improvement of Higher Education Personnel (CAPES) Brazilian government agency. v ABSTRACT Urio, Paulo Roberto (2017). "Complex network component unfolding using a particle competition technique". Master's dissertation. São Carlos: Instituto de Ciências Matemáticas e de Computação (ICMC/USP), 75 This work applies complex network theory to the problem of semi-supervised and unsupervised learning in networks that are representations of multivariate datasets. Complex networks allow the use of nonlinear dynamical systems to represent behaviors according to the connectivity patterns of networks. Inspired by behavior observed in nature, such as competition for limited resources, dynamical system models can be employed to uncover the organizational structure of a network. In this dissertation, we develop a technique for classifying data represented as interaction networks. As part of the technique, we model a dynamical system inspired by the biological dynamics of resource competition. So far, similar methods have focused on vertices as the resource of competition. We introduce edges as the resource of competition. In doing so, the connectivity pattern of a network might be used not only in the dynamical system simulation but in the learning task as well.
de energia de uma usina termoelétrica. Após a aná-lise do algoritmo sequencial, neste trabalho, o PDEE será tratado com um algoritmo paralelo para GPGPUs em CUDA. O algoritmo proposto é uma Evolução Diferencial (ED) utilizando a técnica de ensemble de operadores de mutação. A ED é uma técnica estocástica de otimização baseada em população, desenvolvida para a otimização de valores reais enquanto o ensemble de operadores de mutação permite que várias configurações de parâmetros e estratégias possam ser utilizadas em cada etapa da evolução do algoritmo. Três instâncias de teste, considerando os efeitos de ponto de válvula, são adotadas para verificar a eficiência do método proposto. Os resultados obtidos são favoravelmente comparados com aqueles descritos na literatura da área em termos de qualidade das soluções obtidas. A versão paralela obteve speedups significativos mantendo a boa qualidade das soluções encontradas.Palavras-chave: computação evolucionária, ensemble de operadores de mutação, processamento paralelo.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.