This event has now passed and was part of the Safe and Trusted AI Summer School 2025. Please find recording of the event in the video underneath the bio.
Sequential decision making under uncertainty is a principled approach to planning sequences of decisions in unpredictable, unknown or noisy environments. It has applications to, for example, robotics, autonomous vehicles or medical decision making. These are settings where ensuring system safety can be essential. This tutorial will describe how techniques from formal verification can be used to offer rigorous mathematical guarantees as to the safety, reliability or efficiency of sequential decision making processes. We will describe how techniques from the field of probabilistic model checking can be applied here, and review recent advances in this area that provide support for multi-agent and data-driven decision making.