Reinforcement Learning with Guarantees that Hold for Ever

Ernst Moritz Hahn; Mateo Perez; Sven Schewe; Fabio Somenzi; Ashutosh Trivedi; Dominik Wojtczak

doi:10.1007/978-3-031-15008-1_1

Reinforcement Learning with Guarantees that Hold for Ever

Ernst Moritz Hahn, Mateo Perez, Sven Schewe^*, Fabio Somenzi, Ashutosh Trivedi, Dominik Wojtczak

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

53 Downloads (Pure)

Abstract

Reinforcement learning is a successful explore-and-exploit approach, where a controller tries to learn how to navigate an unknown environment. The principle approach is for an intelligent agent to learn how to maximise expected rewards. But what happens if the objective refers to non-terminating systems? We can obviously not wait until an infinite amount of time has passed, assess the success, and update. But what can we do? This talk will tell.

Original language	English
Title of host publication	Formal Methods for Industrial Critical Systems
Subtitle of host publication	27th International Conference, FMICS 2022, Warsaw, Poland, September 14-15, 2022, Proceedings
Editors	Jan Friso Groote, Marieke Huisman
Place of Publication	Cham
Publisher	Springer
Pages	3-7
Number of pages	5
ISBN (Electronic)	978-3-031-15008-1
ISBN (Print)	978-3-031-15007-4
DOIs	https://doi.org/10.1007/978-3-031-15008-1_1
Publication status	Published - 5 Sept 2022

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	13487
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Keywords

2023 OA procedure

Access to Document

10.1007/978-3-031-15008-1_1

ArticleFinal published version, 162 KBLicence: Taverne

Cite this

Hahn, E. M., Perez, M., Schewe, S., Somenzi, F., Trivedi, A., & Wojtczak, D. (2022). Reinforcement Learning with Guarantees that Hold for Ever. In J. F. Groote, & M. Huisman (Eds.), Formal Methods for Industrial Critical Systems: 27th International Conference, FMICS 2022, Warsaw, Poland, September 14-15, 2022, Proceedings (pp. 3-7). (Lecture Notes in Computer Science; Vol. 13487). Springer. https://doi.org/10.1007/978-3-031-15008-1_1

Hahn, Ernst Moritz ; Perez, Mateo ; Schewe, Sven et al. / Reinforcement Learning with Guarantees that Hold for Ever. Formal Methods for Industrial Critical Systems: 27th International Conference, FMICS 2022, Warsaw, Poland, September 14-15, 2022, Proceedings. editor / Jan Friso Groote ; Marieke Huisman. Cham : Springer, 2022. pp. 3-7 (Lecture Notes in Computer Science).

@inproceedings{ce3a34ab46714e8e85f5d60b28ed021d,

title = "Reinforcement Learning with Guarantees that Hold for Ever",

abstract = "Reinforcement learning is a successful explore-and-exploit approach, where a controller tries to learn how to navigate an unknown environment. The principle approach is for an intelligent agent to learn how to maximise expected rewards. But what happens if the objective refers to non-terminating systems? We can obviously not wait until an infinite amount of time has passed, assess the success, and update. But what can we do? This talk will tell.",

keywords = "2023 OA procedure",

author = "Hahn, {Ernst Moritz} and Mateo Perez and Sven Schewe and Fabio Somenzi and Ashutosh Trivedi and Dominik Wojtczak",

year = "2022",

month = sep,

day = "5",

doi = "10.1007/978-3-031-15008-1_1",

language = "English",

isbn = "978-3-031-15007-4",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "3--7",

editor = "Groote, {Jan Friso} and Marieke Huisman",

booktitle = "Formal Methods for Industrial Critical Systems",

address = "Germany",

}

Hahn, EM, Perez, M, Schewe, S, Somenzi, F, Trivedi, A & Wojtczak, D 2022, Reinforcement Learning with Guarantees that Hold for Ever. in JF Groote & M Huisman (eds), Formal Methods for Industrial Critical Systems: 27th International Conference, FMICS 2022, Warsaw, Poland, September 14-15, 2022, Proceedings. Lecture Notes in Computer Science, vol. 13487, Springer, Cham, pp. 3-7. https://doi.org/10.1007/978-3-031-15008-1_1

Reinforcement Learning with Guarantees that Hold for Ever. / Hahn, Ernst Moritz; Perez, Mateo; Schewe, Sven et al.
Formal Methods for Industrial Critical Systems: 27th International Conference, FMICS 2022, Warsaw, Poland, September 14-15, 2022, Proceedings. ed. / Jan Friso Groote; Marieke Huisman. Cham: Springer, 2022. p. 3-7 (Lecture Notes in Computer Science; Vol. 13487).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Reinforcement Learning with Guarantees that Hold for Ever

AU - Hahn, Ernst Moritz

AU - Perez, Mateo

AU - Schewe, Sven

AU - Somenzi, Fabio

AU - Trivedi, Ashutosh

AU - Wojtczak, Dominik

PY - 2022/9/5

Y1 - 2022/9/5

N2 - Reinforcement learning is a successful explore-and-exploit approach, where a controller tries to learn how to navigate an unknown environment. The principle approach is for an intelligent agent to learn how to maximise expected rewards. But what happens if the objective refers to non-terminating systems? We can obviously not wait until an infinite amount of time has passed, assess the success, and update. But what can we do? This talk will tell.

AB - Reinforcement learning is a successful explore-and-exploit approach, where a controller tries to learn how to navigate an unknown environment. The principle approach is for an intelligent agent to learn how to maximise expected rewards. But what happens if the objective refers to non-terminating systems? We can obviously not wait until an infinite amount of time has passed, assess the success, and update. But what can we do? This talk will tell.

KW - 2023 OA procedure

U2 - 10.1007/978-3-031-15008-1_1

DO - 10.1007/978-3-031-15008-1_1

M3 - Conference contribution

SN - 978-3-031-15007-4

T3 - Lecture Notes in Computer Science

SP - 3

EP - 7

BT - Formal Methods for Industrial Critical Systems

A2 - Groote, Jan Friso

A2 - Huisman, Marieke

PB - Springer

CY - Cham

ER -

Hahn EM, Perez M, Schewe S, Somenzi F, Trivedi A, Wojtczak D. Reinforcement Learning with Guarantees that Hold for Ever. In Groote JF, Huisman M, editors, Formal Methods for Industrial Critical Systems: 27th International Conference, FMICS 2022, Warsaw, Poland, September 14-15, 2022, Proceedings. Cham: Springer. 2022. p. 3-7. (Lecture Notes in Computer Science). doi: 10.1007/978-3-031-15008-1_1

Reinforcement Learning with Guarantees that Hold for Ever

Abstract

Publication series

Keywords

Access to Document

Fingerprint

Cite this