Internal representations analysis in spiking neural networks

Detalhes bibliográficos
Ano de defesa: 2024
Autor(a) principal: de Sant\'Ana, João Henrique
Orientador(a): Não Informado pela instituição
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: eng
Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
Programa de Pós-Graduação: Não Informado pela instituição
Departamento: Não Informado pela instituição
País: Não Informado pela instituição
Palavras-chave em Português:
Link de acesso: https://www.teses.usp.br/teses/disponiveis/43/43134/tde-17062024-144329/
Resumo: Optimal information processing in peripheral sensory systems has been associated in several examples to the signature of a critical or near critical state. Furthermore, cortical systems have also been described to be in a critical state in both wake and anesthetized experimental models, both in vitro and in vivo. Using a methodology inspired in the biological setup, we investigate whether a similar signature characterizes the internal representations (IR) of a multilayer (deep) spiking artificial neural network performing a recognition task. The increase of the characteristic time of the decay of the correlation of fluctuations of the IR, found when the network input changes are indications of a broad-tailed distribution of IR fluctuations. The broad tails are present even when the network is not yet capable of performing the classification tasks, either due to partial training or to the effect of a low dose of anesthesia in a simple model. This is a signature only for significant changes involving labeled inputs. However, we don\'t find enough evidence of power law distributions of avalanche size and duration. Finally, the perturbational complexity index (PCI) of IR was measured. The PCI distinguished levels of training time and also effect of anesthesia leading to a characterization of the spatio-temporal pattern of activity in the internal layers.