User profiles for Jonathan Ferrer-Mestres

Jonathan Ferrer-Mestres

Lead Machine Learning Applied Scientist - Inception Institute of Artificial Intelligence, G42
Verified email at core42.ai
Cited by 151

A primer on partially observable Markov decision processes (POMDPs)

…, S Nicol, CS Fletcher, J FerrerMestres - Methods in Ecology …, 2021 - Wiley Online Library
Partially observable Markov decision processes (POMDPs) are a convenient mathematical
model to solve sequential decision‐making problems under imperfect observations. Most …

Combined task and motion planning as classical AI planning

J Ferrer-Mestres, G Frances, H Geffner - arXiv preprint arXiv:1706.06927, 2017 - arxiv.org
Planning in robotics is often split into task and motion planning. The high-level, symbolic
task planner decides what needs to be done, while the motion planner checks feasibility and …

Synthesis of Procedural Models for Deterministic Transition Systems

J Segovia-Aguas, J Ferrer-Mestres, S Jiménez - ECAI 2023, 2023 - ebooks.iospress.nl
This paper introduces a general approach for synthesizing procedural models of the state-transitions
of a given discrete system. The approach is general in that it accepts different target …

Interpretable Solutions for Stochastic Dynamic Programming

J Ferrer-Mestres, TG Dietterich, O Buffet, I Chades - bioRxiv, 2024 - biorxiv.org
In conservation of biodiversity, natural resource management and behavioural ecology,
stochastic dynamic programming, and its mathematical framework, Markov decision processes (…

KN-MOMDPs: Towards interpretable solutions for adaptive management

J Ferrer-Mestres, TG Dietterich, O Buffet… - Proceedings of the AAAI …, 2021 - ojs.aaai.org
In biodiversity conservation, adaptive management (AM) is the principal tool for decision
making under uncertainty. AM problems are planning problems that can be modelled using …

[PDF][PDF] Planning with state constraints and its application to combined task and motion planning

J Ferrer-Mestres, G Frances… - Proceedings of the ICAPS …, 2015 - gfrances.github.io
Most of the key computational ideas in planning have been developed for planning languages
where action preconditions and goals are conjunctions of propositional atoms. These …

Solving k-mdps

J Ferrer-Mestres, TG Dietterich, O Buffet… - Proceedings of the …, 2020 - aaai.org
Markov Decision Processes (MDPs) are employed to model sequential decision-making
problems under uncertainty. Traditionally, algorithms to solve MDPs have focused on solving …

TALKS: A systematic framework for resolving model-data discrepancies

MP Vilas, F Egger, MP Adams, HR Maier… - … Modelling & Software, 2023 - Elsevier
Abstract Models and data play an important role in informing decision-making in environmental
systems, providing different and complementary information. Multiple frameworks have …

Path-finding in dynamic environments with PDDL-planners

V Estivill-Castro, J Ferrer-Mestres - 2013 16th International …, 2013 - ieeexplore.ieee.org
The standardization of planning problems by their descriptions in the PDDL has resulted in
clear benchmarking of planners, and thus, in significant advances in reliable and efficient …

Planning with partially specified behaviors

J Segovia-Aguas, J Ferrer-Mestres… - Artificial Intelligence …, 2016 - ebooks.iospress.nl
In this paper we present a framework called Planning with Partially Specified Behaviors, or
PPSB, for combining reinforcement learning and planning to solve sequential decision …