Backpropagation
Vytvořeno: 30. 10. 2017 od Architektura III

Workshop by Shota Tsikoliya for studio Architecture III.

Backpropagation is a method used in artificial neural networks to calculate the error contribution of each neuron after a batch of data is processed. During the workshop participants will use supervised machine learning to generate a 2D gradient between various geometries produced via parametric model. The approach will potentially create non-obvious design solutions of
a philharmonic hall.