RT Generic
T1 Graph grammars for complex behavior
T2 Gramática de grafos para comportamiento complejo
A1 Andrés Martínez, Pablo
K1 Grafos, Teoría de
K1 Redes neuronales (Informática)
K1 Computación evolutiva
K1 Informática - Trabajos Fin de Grado
K1 Grado en Ingeniería Informática - Trabajos Fin de Grado
AB Background activity is the biological phenomenon that prevents the brain of an aliveorganism from reaching a state of complete inactivity. The neuroscientist communityclaims that it is related to cognitive functions such as memory and the exploration ofpreviously sensed experiences.Artificial neural networks were originally developed as a nervous system model. InComputer Science, they have been applied in function approximation and pattern recognitionproblems. However, dynamics of the typically used paradigms are not appropriatefor the replication of processes such as background activity. When the goal is toreproduce the behavior of real neural networks, the most adequate model is the SpikingNeural Network (SNN), whose elements closely resemble the biological neurons.Our objective is to develop an algorithm that generates SNN topologies able to maintainbackground activity. The topology of an SNN is described as a graph, thus, the firstcontribution of this project is a grammar formalism to generate them. That formalism isapplied by an automated search process in order to find SNNs that are able to maintainbackground activity. This search is done by an evolutionary algorithm, which developsa population of SNNs and applies successive transformations to them, gradually increasingtheir ability to fulfill the proposed objective.Considering that the different SNNs of the population are independent of each other,the time required to execute the algorithm can be noticeably reduced when using parallelcomputation. In order to obtain the results discussed in this document, the programwas run over 40 cores of the local supercomputing node, which is part of the SpanishSupercomputing Network. The resulting execution time is decreased in an order ofmagnitude compared to the one that would be required in a quad-core personal computer.This was crucial for the development of the project, as it considerably improvedour ability to manage the process of obtaining and studying the results.
YR 2017
FD 2017-02-10
LK http://hdl.handle.net/10630/12983
UL http://hdl.handle.net/10630/12983
LA spa
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 14 abr 2026