Bio

I received my AA at Tallahassee Community College and plan to attend Florida State University for degrees in physics and computational science in the fall. I hope to pursue a PhD in theoretical physics and develop an open-source exploratory n-body simulation software. I have been programming various simulations since around age 8 or 9 and applied this early experience at the University of West Florida, where I simulated the particle-like interactions between quantum vortices in type I and II superconductors. I have major interests in digital rights, GNU/Linux systems, augmenting neural networks with cellular automata and improving computer algebra systems.

Project: Machine Learning with Boolean Networks

We will entertain the concept of constructing the analog of a feedforward neural network [1] composed entirely out of Boolean functions [2], each of which accepts two inputs. A greedy Monte Carlo [3, 4] approach is utilized for training the network with neuron randomization selection based on a truncated normal distribution centered at the center-most layer of the network.

References

[1] Wikipedia. “Feedforward Neural Network.” (Sep 14, 2016) en.wikipedia.org/wiki/Feedforward_neural_network.

[2] Wikipedia. “Boolean Network.” (Sep 14, 2016) en.wikipedia.org/wiki/Boolean_network.

[3] Wikipedia. “Monte Carlo Method.” (Sep 14, 2016) en.wikipedia.org/wiki/Monte_Carlo_method.

[4] Wikipedia. “Greedy Algorithm.” (Sep 14, 2016) en.wikipedia.org/wiki/Greedy_algorithm.

[5] Wikipedia. “Field-Programmable Gate Array.” (Sep 14, 2016) en.wikipedia.org/wiki/Field-programmable_gate _array.

Favorite 3-Color 2D Totalistic Cellular Automaton

Rule 656599508

I picked this rule due to its similarity to Conway’s Game of Life. Complex structures can be seen emerging from random initial conditions. In general, these structures appear much larger in size than similar forms found in outer totalistic rule 224.