After completing his Ph.D. in physics, Declan focused more on his computer science related interests. He started a small software and optimization consulting business with several friends, was awarded a patent through some of that work, did some mathematical modeling consulting, and published a machine learning conference article on other work.
“I believe for AI to do more complex tasks, it will inevitably need to build and use models of the world it interacts with. This is what I'm focusing on currently; learning a model of the environment that our agents will then use to plan future actions. Ideally, this could lead to both higher efficiency than model-free methods as well as more optimal solutions.”
Outracing Champion Gran Turismo Drivers with Deep Reinforcement Learning
Many potential applications of artificial intelligence involve making real-time decisions in physical systems while interacting with humans. Automobile racing represents an extreme example of these conditions; drivers must execute complex tactical manoeuvres to pass or block…