Differences

This shows you the differences between two versions of the page.

--- people:taylor_berger:infinite_brain_paper [2014/10/06 23:00] – created tberge01
+++ people:taylor_berger:infinite_brain_paper [2014/10/20 18:35] (current) – tberge01
@@ Line 1: / Line 1: @@
-===== An Indefinite Brain: Simplifying Neural Networks in a Robust Computational System =====
+===== An Indefinitely Scalable Brain: Implicit Neural Networks in a Spatially Distributed System =====
-//Abstract: // Current models of neural networks replace the spacial features of neuronal systems in favor of a mathematical description. This paper forgoes the mathematical model in favor of a more robust, spatially distributed neural network and shows pattern recognition is not only viable, but can be reconstructed in a volatile system. We show that this type of neural network can be scaled indefinitely and learns arbitrary patterns despite adverse learning conditions that would break traditional mathematical models of neural networks.
+//Abstract: // ...(problem statement)... We define an implicit, spatially distributed neural network and show pattern recognition is not only viable, but robust in its classification tasks in a volatile system. We use the Moveable Feast Machine architecture to investigate a neural network with implied connections between neurons based on their proximity. We show that this type of neural network can be scaled indefinitely and learns effectively despite adverse learning conditions. We show this type of neural network is capable of identifying patterns and performs better than making a random decision in a two-class classification task.