Review of Unsupervised Adaptive Resonance Theory

Tripathi, Agraj; Srivastava, Geetika; Singh, K.K.; Maurya, Pradeep Kumar

doi:10.1109/confluence.2019.8776941

Cited by 3 publications

(2 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Of course, as nature inspires engineers, work on the brain would provide new ideas to develop intelligent agents and systems. New approaches in machine learning already are due to work on the brain as Barto et al's reinforcement learning [ 5 , 6 ] and Grossberg's adaptive resonance theory [ 7 ]. The spiking neuron models are also considered in developing new machine learning methods [ 8 ], and working on the spiking neural network models would improve these approaches.…”

Section: Introductionmentioning

confidence: 99%

“…Using this model, we investigated the effect of changing the connection weights on synchronization. It is claimed that the relation between synchrony and the neuronal code aroused from the activation of a group of neurons will inspire new learning techniques [ 7 ]. This model can be used also to study the role of short-term plasticity and STDP (spike timing-dependent plasticity) on learning as in [ 5 , 19 ], where the Brian simulator has a library that is related to STDP, by defining the pre- and postsynaptic neuronal groups, which can be applied to our model [ 20 ].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Computational Model of the Brain Cortex and Its Synchronization

Kbah

2020

BioMed Research International

View full text Add to dashboard Cite

Obtaining the computational models for the functioning of the brain gives us a chance to understand the brain functionality thoroughly. This would help the development of better treatments for neurological illnesses and disorders. We created a cortical model using Python language using the Brian simulator. The Brian simulator is specialized in simulating the neuronal connections and synaptic interconnections. The dynamic connection model has multiple parameters in order to ensure an accurate simulation (Bowman, 2016). We concentrated on the connection weights and studied their effect on the interactivity and connectivity of the cortical neurons in the same cortical layer and across multiple layers. As synchronization helps us to measure the degree of correlation between two or more neuronal groups, the synchronization between the neuronal groups, which are connected across layers, is considered. Despite its obvious importance, there are no sufficient studies concerned about the synchronization in the simulated cortical models. Such studies can help in examining the hypothesis and the dynamical behavior of the simulated model. In this paper, we simulated a cortical model and dynamical behavior and then studied the effect of input noise on its internal neuronal networks and their synchronization.

show abstract

Section: Introductionmentioning

confidence: 99%