What is reinforcement learning?

RL is a form of machine learning where an agent learns from a reward signal that it receives after performing a sequence of actions. Imagine a chess player or a player of a video game. This technique was the reason for the success of AlphaGo, the first AI to beat the world champion in the game of GO. People say it's the next step towards general artificial intelligence.   

Why a course on applied reinforcement learning?

The real value of RL lies not in playing games, but in solving real-world sequential decision making problems. These problems exist everywhere, in every company and in every live system. We are now at a point where we have the tools and computational resources to tackle these challenges, which can bring huge benefits to all these fields. I aim to enable practitioners to find solutions for the concrete problems they face. I expect that in the future, nearly all data science will be RL-based.

What will participants learn in your course?

We will start with an overview of the important classical approaches to RL problems. Then we will focus mainly on deep RL and cover the most powerful RL algorithms. An important topic will be different ways to increase sample efficiency. And we will always emphasize the practical aspects of developing RL solutions.

What is the value for companies to employ RL?

The value is immense, much bigger than what data science is commonly used for today. As digitization and automation is advancing, more data will be available, and it will become easier to build models to simulate core business processes. If you want to know more about this, you can read my recent article on this topic here.

How is the course structured?

The course will be held in a hybrid format, so it is possible to participate in person or online. There will be some practical exercise for each topic, so participants will not only learn the theory, but also how to implement things by themselves. We will also form small teams to work on a project together, which makes learning more fun and gives participants the joy of achieving something together.

Who should attend this course?

The course is for everyone curious about the exciting new field of RL solutions. As a prerequisite, participants should have some experience with machine learning and have created at least one neural network model.

The course starts on August 26 and takes place in the evenings. The application deadline is August 12. More information and registration: Applied Reinforcement Learning

Learn more about the wide range of continuing education courses offered by the Institute of Computational Life Sciences in the field of Computational Science and Artificial Intelligence: www.zhaw.ch/icls/continuingeducation