The aim of this project is to design simulation-based processes for creating and evaluating Bayesian Networks that can be applied to real-world problems, such as the problem of the reference class, problem of the priors, causality, and others - or to show limitations of such an approach. The process should be hybrid, in that it plays to the strengths of both the BN and human modellers and users (arguers) by combining probabilistic updating rules with rational deliberation to decide which events, causalities and probabilities to apply.

• Ludi van Leeuwen; MSc. (RUG)
• prof.dr. Bart Verheij (RUG)
• prof.dr. Rineke Verbrugge (RUG)

Our aim is to design systems that allow for monitoring AI systems in order to check if the AI system adheres to various types of constraint, like norms or protocols. These constraint are the result of the context the AI system operates in and may be the result of limited resources, laws and regulations, ethical or societal considerations. The monitoring system will model these constraints in a human-intuitive and transparent way in order to facilitate updating the constraints by human stakeholders, as well as to allow for explaining which constraints the AI system is or isn't adhering to, thereby facilitating the communication and collaboration between human and AI systems.

• prof. dr. M.M. (Mehdi) Dastani
• A.T. (Annet) Onnes MSc

• dr.ir. R.I.J. (Roel) Dobbe (TU Delft)

By investigating the relations between knowledge, reasoning and data, we aim to develop mechanisms for the verification and evaluation of hybrid systems that combine manual knowledge-based design and learning from data. The focus will be on structures used in reasoning and decision-making, in particular logical and probabilistic relations (as in Bayesian networks) and reasons (pro and con) and exceptions (as in argument-based decision making).

In Responsible HI, there is a need to ensure that the behaviour of HI systems is aligned with legal and moral considerations. So the input and output of a system must meet a given set of principles. Verification, evaluation and interaction mechanisms are needed that ensure such alignment, that show to what extent alignment has been achieved, and that help improve the alignment. 

• Cor Steging (RUG)
• Bart Verheij (RUG)

Bayesian networks (BNs) provide decision support in complex investigative domains where uncertainty plays a role, such as medicine, forensics and risk assessment. Yet, BNs are only sparsely used in practice. In data-poor domains, they have to be manually constructed, which is too time-consuming to support pressing decisions. Furthermore, few domain experts have the mathematical background to build a BN, a graph representing dependencies among variables with probability distributions over these variables. So despite the increased analytical power a BN could bring with respect to, for example, evidence aggregation or sensitivity analysis, many experts still use more qualitative concepts such as scenarios (stories, cases, timelines) and arguments (evidence graphs, ordered lists), which convey verbally expressed uncertainty ("strong evidence", "plausible scenarios").

If BNs are to be used in actual investigations, we need software tools and interfaces for BN construction that are engineered into the heart of the decision-making process. These tools should be based on familiar, more linguistically-oriented concepts such as arguments and stories, and complemented by algorithms intended to speed up and facilitate the BN-building process.

• dr. Remi Wieten
• prof. dr. mr. H. (Henry) Prakken
• prof. dr. F.J. (Floris) Bex

Lucia de Berk found out first-hand: evidence based on statistics can easily lead to errors. This project aims to help prevent this sort of error from occurring. The project's new approach is to link the successful statistical modelling technique of Bayesian networks to models that effectively dovetail legal argumentation and scenario construction in the legal world.

This project consists of two sub-projects. In one project (phd-student: Sjoerd Timmer, UU) the goal is to connect Bayesian networks with argumentation approaches; in the other project (phd-student Charlotte Vlek, RUG) the focus is on connecting Bayesian networks with scenario- or story-based methods. I am a co-promotor for both phd projects.

• prof. dr. mr. H. (Henry) Prakken
• prof. dr. J.J.C. (John-Jules) Meyer

• S.T. Timmer
• C.S. Vlek MSc
• dr. B. Verheij
• prof. dr. L.C. Verbrugge (Department of Artificial Intelligence; University of Groningen)