FEAR 2025 

This tutorial introduces the object of interest and methodology of the field in broad strokes. It offers an overview of the filed developments including most impactful references, and presents the open challenges. This tutorial is perfect for you who is considering doing research in machine ethics. No background is necessary.

It is not enough that machines behave ethically. We also need to be able to prove that they do. This tutorial focuses on formal verification for ethical behaviour. It lightly introduces formal verification and the dives into the specific developments and challenges of the field. A background in formal verification is appreciated but for those with a curious disposition, the tutorial is available even without it.

Autonomous agents based on reinforcement learning have attracted a lot of interest. What a reinforcement learning agent chooses to do is governed by its reward function. This tutorial introduces an approach to specifying reward functions based on "reward machines". Reward machines allow the specification of rewards based on the history of an agent's interaction with its environment. As such they can be used to specify norms an agent should conform to while learning to perform another task. After briefly introducing reinforcement learning and reward machines, we will discuss how reward machines can be used to learn to comply with norms in a multi-agent setting. Some basic knowledge of reinforcement learning and some familiarity with formal methods is desirable but not essential.