Hello,

I am a PhD candidate in Artificial Intelligence at Utrecht University. I am looking into how to monitor black-box AI systems and teach various courses in AI and CS. I am interested in Responsible AI, normative monitoring, Bayesian Networks, knowledge-driven AI, and ethics of AI.

Scroll down to find out more about what I do and have done.

a.t.onnes[at]uu.nl.

Doctoral research

I started my PhD on "Monitoring Black-box AI systems using Bayesian Networks" in 2020 and am supervised by: Silja Renooij, Mehdi Dastani and Roel Dobbe (TUD)

Monitoring Black-box AI systems using Bayesian Networks

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.

For my implementations I work with and utilise the advantages of Bayesian Networks and consider work on systems safety and AI and risk analysis.

Publications

  • Onnes, A.: Monitoring AI systems: A problem analysis, framework and outlook. In: Proceedings of the First International Conference on Hybrid-Artificial Intelligence. Frontiers in Artificial Intelligence and Applications, vol. 354, pp. 238–240 (2022)
  • Onnes, A., Dastani, M., Renooij, S.: Bayesian network conflict detection for normative monitoring of black-box systems. In: Proceedings of the Thirty-Sixth FLAIRS Conference. vol. 36. Florida Online Journals (2023)
  • Onnes, A., Dastani, M., Renooij, S.: Normative Monitoring Using Bayesian Networks: Defining a Threshold for Conflict Detection. (In press, accepted for ECSQARU 2023)

Also presented my work on Bayesian network conflict detection for normative monitoring of black-box systems at Workshop on the Design of Responsible Hybrid Intelligence during HHAI2023

Supervision

My daily supervisor and co-promotor at Utrecht University is Silja Renooij and my promotor is Mehdi Dastani. As part of the Hyrbid Intelligence consortium I also have external supervisor, this is Roel Dobbe.

Curriculum Vitae

Teaching

Utrecht University

  • Seminar teacher for Philosophy of AI in 2021, 2022 and 2023
  • Co-lecturer for Intelligent Systems in2022 and 2023
  • Assistant for AI for gametechnology 2020

The University of Edinburgh

  • TA for the Data ethics, AI and responsible innovation MOOC

University of Groningen

  • Various TA appointments between 2014 and 2018 for Introduction to Logic, Advanced Logic, Autonomous Systems, Webdesign, General Linguistics and Speech and Language Technology

Further activities during my PhD

  • 2021 - Now Education Advisory Committee Bachelor Computing Science at Utrecht University
  • 2020 - 2022 Organisation Hybrid Intelligence PhD reading group and colloquia
  • 2023 - Now Hybrid Intelligence diversity committee

Education

  • 2018 - 2019 MSc Mind, Language and Embodied Cognition (The University of Edinburgh)
  • 2014 - 2018 BA Philosophy of a Specific Discipline (University of Groningen)
  • 2013 - 2018 BSc Artificial Intelligence (University of Groningen)

About me

You can download a more extensive (but probably less up-to-date) CV here.

You can also find me on Linkedin or (not recommended) Twitter.