Evolving Self-Driving Game AI and Self-Destructing Rockets using Genetic Algorithms, Neural Networks

Folkets Hus

Genetic algorithms can be used in machine learning to evolve optimal solutions to challenges through learning and experimentation. Neural networks are used to perform complex tasks such as image recognition and autonomous robots. Combining these two technologies can create a machine learning system that can learn to drive a car in a driving game or land a rocket booster in a simulation with no prior knowledge.

In this demo intensive session Alan will demonstrate how a basic genetic algorithm can be implemented in C# to evolve and optimize the simulated landing of rocket boosters. He will then introduce a simple C# based neural network that can control more complex systems, such as a self-learning AI in a driving game. He will share his thoughts on how to get to grips with the basics of machine learning, and discus the futures of ML technology on the Azure platform.