Planning for potential: efficient safe reinforcement learning

Autor: Floris den Hengst, Vincent François-Lavet, Mark Hoogendoorn, Frank van Harmelen

Publikováno v: den Hengst, F, François-Lavet, V, Hoogendoorn, M & van Harmelen, F 2022, ' Planning for potential: efficient safe reinforcement learning ', Machine Learning, vol. 111, no. 6, pp. 2255-2274 . https://doi.org/10.1007/s10994-022-06143-6
Machine Learning, 111(6), 2255-2274. Springer Netherlands

Deep reinforcement learning (DRL) has shown remarkable success in artificial domains and in some real-world applications. However, substantial challenges remain such as learning efficiently under safety constraints. Adherence to safety constraints is

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::26535ad5c5469040724c026ef002bc96
https://research.vu.nl/en/publications/fa7580e0-f14c-464a-804f-e0c72d01c56a

Reinforcement Learning for Radiotherapy Dose Fractioning Automation

Autor: Paul Desbordes, Grégoire Moreau, Benoît Macq, Vincent François-Lavet

Publikováno v: Biomedicines, Vol 9, Iss 214, p 214 (2021)
Biomedicines
Volume 9
Issue 2
Biomedicines, Vol. 9, no.2, p. 214 (2021)

External beam radiotherapy cancer treatment aims to deliver dose fractions to slowly destroy a tumor while avoiding severe side effects in surrounding healthy tissues. To automate the dose fraction schedules, this paper investigates how deep reinforc

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d91094791671e9e28eb4efd30a5a75a5
https://www.mdpi.com/2227-9059/9/2/214

On Overfitting and Asymptotic Bias in Batch Reinforcement Learning with Partial Observability (Extended Abstract)

Autor: Damien Ernst, Vincent François-Lavet, Joelle Pineau, Guillaume Rabusseau, Raphael Fonteneau

Publikováno v: IJCAI

When an agent has limited information on its environment, the suboptimality of an RL algorithm can be decomposed into the sum of two terms: a term related to an asymptotic bias (suboptimality with unlimited data) and a term due to overfitting (additi

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::127b84816ddd692bf9804c66dce8b90f
https://doi.org/10.24963/ijcai.2020/706

Combined Reinforcement Learning via Abstract Representations

Autor: Joelle Pineau, Vincent François-Lavet, Yoshua Bengio, Doina Precup

Publikováno v: AAAI

In the quest for efficient and robust reinforcement learning methods, both model-free and model-based approaches offer advantages. In this paper we propose a new way of explicitly bridging both approaches via a shared low-dimensional learned encoding

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ba07fc0fe68863bff3c12bdace2e8ee6
http://arxiv.org/abs/1809.04506

An Introduction to Deep Reinforcement Learning

Autor: Vincent François-Lavet, Peter Henderson, Riashat Islam, Marc G. Bellemare, Joelle Pineau

Deep reinforcement learning is the combination of reinforcement learning (RL) and deep learning. This field of research has recently been able to solve a wide range of complex decision-making tasks that were previously out of reach for a machine. Dee