An Evolving Artificial Neural Network for the Investigation of Rat Exploratory Behavior

Costa, Ariadne de Andrade; Tinós, Renato

doi:10.1109/bracis.2014.29

Cited by 3 publications

(4 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These amounts are considerably lower than the 4 hidden neurons on the static network of Model 1, reducing the network complexity and computational costs. Comparing these results to the evolved ANNs obtained for the LCE [Costa and Tinos 2014], the light-dark box seems to be a simpler apparatus to navigate on, since it requires less elements in the ANN. The mean number of sensors found for the LCE was 2.4, accompanied by 3 hidden neurons (it is worth mentioning that the mean was calculated over all the rats per generation, not only over the best scored ones).…”

Section: Architectures Evolved By the Ga For Modelmentioning

confidence: 76%

“…The luminosity sensor with the highest incidence was the "LR", present in 14 (46.67%) of the computational agents. Apparently, side sensors are more useful than frontal ones (also in the LCE [Costa and Tinos 2014]).…”

Section: Architectures Evolved By the Ga For Modelmentioning

confidence: 99%

“…In the first one, only the weights of the ANNs are evolved, while in the second the network architecture is also optimized. Something similar was done for the LCE [Costa and Tinos 2014], but the analysis performed was different from the one presented here. There, the evolution of the average number of sensors and hidden neurons along the generations of the GA was shown, and it were not presented the specific combinations of sensors and hidden neurons evolved, which are shown in the present paper.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Simulation of Rat Behavior in a Light-Dark Box via Neuroevolution

Souza¹,

Silva²,

Tarrega³

et al. 2022

Anais Do XIX Encontro Nacional De Inteligência Artificial E Computacional (ENIAC 2022)

View full text Add to dashboard Cite

The light-dark box is a widely used test for the investigation of animal behavior commonly used to identify and study anxious-like behavioral patterns in rodents. We propose a neuroevolution model for virtual rats in a simulated light-dark box. The virtual rat is controlled by an artificial neural network (ANN) optimized by a genetic algorithm (GA). The fitness function is given by a weighed sum of two terms (punishment and reward). By changing the weight of the punishment term, we are able to simulate the effects of anxiolytic/anxiogenic drugs on rats. We also propose using GAs to optimize the number of the ANN hidden neurons and sensors for the virtual rat. According to the experiments, the best results are obtained by ANNs combining both luminosity and wall sensors.

show abstract

Section: Architectures Evolved By the Ga For Modelmentioning

confidence: 76%

Section: Architectures Evolved By the Ga For Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Simulation of Rat Behavior in a Light-Dark Box via Neuroevolution

Souza¹,

Silva²,

Tarrega³

et al. 2022

Anais Do XIX Encontro Nacional De Inteligência Artificial E Computacional (ENIAC 2022)

View full text Add to dashboard Cite

show abstract

“…Curiosamente, dentre todos os trabalhos publicados envolvendo ratos no LCE, apenas nove exploraram modelos computacionais, de acordo com o conhecimento da autora [18,19,20,21,22,23,24,25,26] [67]. Os elementos que atuam na população, por exemplo selecionando-a ou modificando-a, são chamados de operadores.…”

Section: Estrutura Da Teseunclassified