Aimed at decision makers and scientific or engineering personnel such as data scientists, software engineers, researchers or research managers who would like to acquire the know-how and skills of state-of-the-art methods to solve decision making problems in order to develop their own RL applications.

Prerequisites for this course: This is an advanced course which assumes you have some programming experience in python and fundamental knowledge of machine learning and deep learning (see our dedicated courses).

This course enables the participants to acquire an understanding of the state-of-the-art methods in deep reinforcement learning and the coding skills to start developing RL projects on their own.

The participants will:

• Learn what kind of problems can be solved with reinforcement learning
• Understand the key concepts of RL solutions
• Learn how to build their own RL agents in Python
• Get to know a wide range of RL algorithms
• Learn how to approach their own RL projects

• Where Reinforcement Learning can help
• Theoretical Foundations of RL
• Basic Algorithms for Decision Making
• Algorithms for Model-Free RL and Model-Based RL
• Creating Simulations and RL Environments
• Practical Aspects of developing RL solutions

The course is strongly based on applied exercises. Lectures are split in roughly equal parts of theory and practical programming exercises.

In addition to traditional lectures, the students will experience:

• Self-study (preparation and follow-up)
• Group work
• Parctical programming exercises
• A final project where a specific use case is developed end-to-end 

There are 6 lessons organized once a week on Fridays from 5:30pm to 8pm. The final project presentation will be held during the last lesson.

The first and last event will be held on site at ZHAW near HB Zurich (online participation is possible) to get to know each other, form project groups, facilitate networking and celebrate the project outcomes of the participants. Intermediate events will be held online for higher efficiency and convenience.