Thomas was born April 16 1987 in The Hague, The Netherlands. He was raised in Doetinchem, a small
city in the east of the Netherlands, near the German border. In 2008, he obtained his Bachelor’s degree in Computer Science from the University of Applied Sciences in Arnhem. Following his childhood dream, he subsequently started a job as a videogame programmer in his hometown Doetinchem.
In 2011, Thomas decided to withdraw from the videogame industry and instead try to pursue a career in academia. To this end, he started his Master’s in Computer Science at Utrecht University, on the specialisation track Technical Artificial Intelligence. In 2013 he graduated cum laude from this programme, obtaining his MSc degree. His dissertation, “Mining Exceptional Linear Regression Models with Tree-Constrained Gradient Ascent”, received the runner-up award for Best Master Thesis of Utrecht University’s Faculty of Science.
After his Master’s, he started as a PhD student in the Decision Support Systems group at the department of Information and Computing Sciences of Utrecht University. Unfortunately, external funding issues caused his project to be terminated after around six months. He nevertheless stayed on as a PhD student at Utrecht University after moving to the Intelligent Data Analysis group. Striking up a collaboration with the Imprecise Probability group at Ghent University, in Belgium, Thomas then started working on the theoretical and algorithmic aspects of using continuous-time stochastic processes and the theory of imprecise probabilities, for the robust modelling of dynamical systems under uncertainty.
The UTOPIAE project offered Thomas the opportunity to continue this collaborative research after his contract ended in Utrecht, and in 2017 he started as a PhD student in the Imprecise Probability group at Ghent University, under the supervision of Gert de Cooman and Jasper De Bock. Here he continues his research on imprecise continuous-time Markov chains, while staying in close contact with his previous supervisors in Utrecht. As part of this project, he aims to solve various foundational, theoretical, and algorithmic aspects of robust inference on dynamical systems under different types of uncertainty.
His research interests include uncertainty representation using the theory of imprecise probabilities, reasoning under uncertainty, inference (algorithms) in (imprecise) probabilistic models, (imprecise) stochastic processes, statistical inference, Bayesian statistics, machine learning, data mining, …
Thomas lives in Hoogstraten, Belgium, together with his wife Maartje Kristensen and their two cats.
Research output:
Thomas Krak: Computing Expected Hitting Times for Imprecise Markov Chains, accepted for publication in Proceedings of UQOP 2020, forthcoming.
Natan T’Joens, Thomas Krak, Jasper De Bock, and Gert de Cooman: A Recursive Algorithm for Computing Inferences in Imprecise Markov Chains, in Proceedings of ECSQARU 2019, pp. 455–465, 2019.
Thomas Krak, Natan T’Joens, and Jasper De Bock: Hitting Times and Probabilities for Imprecise Markov Chains, in Proceedings of ISIPTA 2019, pp. 265–275, 2019.
Matthias Troffaes, Thomas Krak, and Henna Bains: Two-State Imprecise Markov Chains for Statistical Modelling of Two-State Non-Markovian Processes, in Proceedings of ISIPTA 2019, pp. 394–403, 2019.
Thomas Krak, Alexander Erreygers, and Jasper De Bock: An Imprecise Probabilistic Estimator for the Transition Rate Matrix of a Continuous-Time Markov Chain, in Proceedings of SMPS 2018, pp.124–132, 2018.
Thomas Krak, Jasper De Bock, and Arno Siebes: Efficient Computation of Updated Lower Expectations for Imprecise Continuous-Time Hidden Markov Chains, in Proceedings of ISIPTA 2017, pp. 193–204, 2017.
Margarita Antoniou, Thomas Krak, Alexander Erreygers, Jasper De Bock, Gert de Cooman, Bounding limit expectations of Markov chains using evolutionary algorithms, OSE, Ljubljana, Slovenia, November 2019
T. Krak: An Introduction to Imprecise Markov Chains, Chapter in Optimization under Uncertainty with Applications to Aerospace Engineering, Springer, 2020.
J. De Bock, T. Krak: Imprecise Markov Chains, Invited Tutorial Session at SMPS 2018, Compiègne, 2018.
T. Krak, C. Greco, T. Basu: A Heuristic Solution to Non-Linear Filtering with Imprecise Probabilities, Abstract & Presentation at UQOP 2019, Paris, 2019.
T. Basu, C. Greco, T. Krak: Imprecise Filtering for Spacecraft Navigation, Presentation at WPMSIIP 2018, Oviedo, 2018.
T. Krak: Learning IP Models using Generalised MAP Estimation, Presentation at WPMSIIP 2017, Compiègne, 2017.
J. De Bock, T. Krak: Imprecise Updating with Continuous Random Variables, Presentation at WPMSIIP 2017, Compiègne, 2017.
T. Krak, J. De Bock: Computing Expected Hitting Times for Imprecise Markov Chains, Abstract & Poster at ISIPTA 2019, Ghent, 2019.