Bioinspire
Country:
Turkey
Year:
2015
Category:
Fulldome
Duration:
0 hours 5 minutes
Director:
Yusuf Emre Kucur, Bahadir Dagdelen/Producers: Void bevoid.co
Age restriction:
General Audience
Language:
No narration

Synopsis

Artificial neural networks(ANNs), a largely used method in machine learning and
cognitive science, are inspired by the biological neural networks, namely the neural
system of the animals. The ANNs are composed of several nodes, layers and
connections which simulate to some extent the message exchange and processing
through a biological neural network. In fact, ANNs tries to find an approximate
functions that evaluate all the inputs in order to give a meaningful output concerning
the data. Each function is specific to the learned data according to which the form of
the ANN is shaped. So, ANN adapt itself to learn from the data, to mine it; it then turns
into a generic processing model that evaluates, classify, categorize the inputs, likewise
our brain, neurons and synapses. As the quantity of layers, nodes and connections
in an artificial neural network increase, the structure gets more complex. Getting the
inspiration from the structure of an ANN, our aim is to redefine the complexity of a
neural network using abstract objects and sound referring to it’s natural form.