MDP

Topology-Driven Solver Selection for Stochastic Shortest Path MDPs via Explainable Machine Learning

Selecting optimal solvers for complex AI tasks grows increasingly difficult as algorithmic options expand. We address this challenge …

Mathieu Gravel, Jaël Champagne Gareau

Résolution efficace de processus décisionnels de Markov par l'exploitation d'approches structurelles et algorithmiques tirant parti de l'architecture moderne des ordinateurs

Cette thèse présente des contributions en planification automatique sous incertitude, un domaine de l’intelligence artificielle. …

Jaël Champagne Gareau

Towards Topologically Diverse Probabilistic Planning Benchmarks

Markov Decision Processes (MDPs) are often used in Artificial Intelligence to solve probabilistic sequential decision-making problems. …

Jaël Champagne Gareau, Éric Beaudry, Vladimir Makarenkov

Cache-Efficient Dynamic Programming MDP Solver

Automated planning research often focuses on developing new algorithms to improve the computational performance of planners, but …

Jaël Champagne Gareau, Guillaume Gosset, Éric Beaudry, Vladimir Makarenkov

Processus Décisionnels de Markov

Ce projet vise à trouver différentes façons d’améliorer la performance des planificateurs de (SSP-)MDP en considérant l’architecture des ordinateurs (p.ex., la mémoire cache, le parallélisme, etc.).

Processus Décisionnels de Markov

Cache-Efficient Memory Representation of Markov Decision Processes

Research in automated planning typically focuses on the development of new or improved algorithms. Yet, an equally important but often …

Jaël Champagne Gareau, Éric Beaudry, Vladimir Makarenkov

pcTVI: Parallel MDP Solver Using a Decomposition Into Independent Chains

Markov Decision Processes (MDPs) are useful to solve real-world probabilistic planning problems. However, finding an optimal solution …

Jaël Champagne Gareau, Éric Beaudry, Vladimir Makarenkov