Neural Networks, Artificial Intellegence, Machine Learning. The word is out and the media is full with them, but what are they and what do they do? Well, I won’t really get into that because there is Google. But I will instead show you what uses for Neural Networks I came up with so far.

Okay, one more (actually 2 more) thing(s):

What excites me about NN’s is that they mimic neuronal constructs, the structures of neurons in a human brain (wich make up your thoughts, your descisions, basically you) And secondly that you can feed them virtually any sort of data and they „learn“, „interpret“, „understand“ it all by themselves.

NN’s indeed feel like the first „human“ computing mechanism. I mean, not really human but imagine placing a person in a new environment. They immeadeatly start trying to make sense of their surroundings, interpret the situation and take according descisions based on what they perceive.

A Neural Network does something very similar. You place it in a new environment by „feeding“ it data (like your eyes are feeding you visual data or your ears feeding you sonic data) and then it goes on and takes the data given and starts forming a complex mathematical model to describe or „make sense“ of the input.

Okay. Here are some of my first attempts of utilizing NN’s, enjoy.

the computer’s face while working

a NN‘ fed with expensive designer clothes

approx. 140.000 iterations

a network learning about galaxies from Hubble

100.000 iterations and counting

a network learning how humans express themselves these days. (it still has quite a lot of mood swings..)

approx. 50.000 iterations

My favourite creation so far: a NN which was fed random images. It couldn’t find any general structures so it started creating abstract art.

approx. 120.000 iterations

This is the current state of my research. I will continue using Neural Networks for creative, non-intuitive purposes and share my results along the way.

For more frequent updates follow me (@timo.matt) on instagram.

_end of excerpt.

Comments are closed.