Software tools

POMDPs with Reveals [code]
Nathanaël Fijalkow, Roman Kniazev, Guillermo A. Pérez, Pierre Vandenhove, 2026. Related paper: Revelations: A Decidable Class of POMDPs with Omega-Regular Objectives, 39th Annual AAAI Conference on Artificial Intelligence (AAAI 2025). Implementation accompanying work on revealing POMDPs, with algorithms for weakly and strongly revealing POMDPs that reduce the analysis to a finite belief-support Markov decision process.
Bolt (Blazingly Fast LTL\(_f\) Learning), with an LTL\(_f\) learning benchmark suite [code] [DOI]
Gabriel Bathie, Nathanaël Fijalkow, Théo Matricon, Baptiste Mouillon, Pierre Vandenhove, 2026. Related paper: LTL\(_f\) Learning Meets Boolean Set Cover, 32nd International Conference on Tools and Algorithms for the Construction and Analysis of Systems (TACAS 2026). Software and benchmark suite for learning LTL\(_f\) specifications via Boolean set cover.
Regular Memory Requirements [code]
Patricia Bouyer, Nathanaël Fijalkow, Mickael Randour, Pierre Vandenhove, 2023. Related paper: How to Play Optimally for Regular Objectives?, 50th EATCS International Colloquium on Automata, Languages and Programming (ICALP 2023). Algorithms that compute minimal memory structures to play optimally in two-player zero-sum games with regular reachability and safety objectives.
Mask R-CNN in OCaml's Owl [code]
Pierre Vandenhove, 2018. Implementation and optimisation of the Mask R-CNN architecture for image segmentation and classification using OCaml's numerical library Owl. Work produced during an internship at OCaml Labs, University of Cambridge. See my internship report for more details.
[Back to main page]