Leibo, JZ;
Perolat, J;
Hughes, E;
Wheelwright, S;
Marblestone, AH;
Duenez-Guzman, E;
Sunehag, P;
... Graepel, T; + view all
(2019)
Malthusian Reinforcement Learning.
In: Elkind, Edith and Veloso, Manuela, (eds.)
Proceedings of the 18th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2019).
(pp. pp. 1099-1107).
IFAAMAS (International Foundation of Autonomous Agents and MultiAgent Systems): Montreal, Canada.
Preview |
Text
Graepel_VoR_pp1099-1107.pdf - Published Version Download (1MB) | Preview |
Abstract
Here we explore a new algorithmic framework for multi-agent reinforcement learning, called Malthusian reinforcement learning, which extends self-play to include fitness-linked population size dynamics that drive ongoing innovation. In Malthusian RL, increases in a subpopulation’s average return drive subsequent increases inits size, just as Thomas Malthus argued in 1798 was the relationship between preindustrial income levels and population growth [24]. Malthusian reinforcement learning harnesses the competitive pressures arising from growing and shrinking population size to drive agents to explore regions of state and policy spaces that they could not otherwise reach. Furthermore, in environments where there are potential gains from specialization and division of labor, we show that Malthusian reinforcement learning is better positioned to take advantage of such synergies than algorithms based on self-play.
Type: | Proceedings paper |
---|---|
Title: | Malthusian Reinforcement Learning |
Event: | 18th International Conference on Autonomous Agents and MultiAgent Systems (AAMAS 2019), 13-17 May 2019, Montreal, Canada |
ISBN: | 978-1-4503-6309-9 |
Open access status: | An open access version is available from UCL Discovery |
Publisher version: | http://www.ifaamas.org/Proceedings/aamas2019/pdfs/... |
Language: | English |
Additional information: | This version is the version of record. For information on re-use, please refer to the publisher’s terms and conditions. |
Keywords: | Intrinsic motivation; Adaptive radiation; Demography; Evolution; Artificial general intelligence |
UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science |
URI: | https://discovery.ucl.ac.uk/id/eprint/10081612 |
Archive Staff Only
![]() |
View Item |