Everything has started from a simple Control of a Robotic Manipulator project in Matlab. I was a junior mechanical engineer, and I was so fascinated by how we can simulate a robot in Matlab and program a controller for it. So much so that I decided to do my master thesis on a more robotics-oriented subject, SLAM. It was there that I got to learn C++, and became interested in computer programming. At the same time, I was developing passion in Control Theory too, as it was heavily based on mathematical proofs, what I have been enjoying since my high school.
Then I moved to South Korea to work at Prof. Yoon's Robots and Intelligent Systems lab, where I worked on Control of Rehabilitation Robots. There, I figured among all aspects of a robotic engineer's job, from designing the robot and choosing mechanical parts, to programming the electronic boards and designing the controller, I, almost always, get hooked to the mathematical interpretation of controllers and the way they can be implemented as a computer program in the most efficient way. Later, I figured there is a whole field about this in Computer Science that people have been working on, it is called Artificial Intelligence. In fact, control theory, robotics, and SLAM are specific problems in this field.
To fulfill my ambitions to be an expert in AI, I joined UNH and started working on my research. Reinforcement Learning, is, in some ways, a continuation of Machine Learning, the science of automatic modeling using data analysis. They are branches of Artificial Intelligence, and help us design algorithms that solve very complicated problems computationally, that human being is not able to even get close to solving it by hand.