TY - GEN
T1 - Batch reinforcement learning on a RoboCup Small Size League keepaway strategy learning problem
AU - Ollino, Franco
AU - Solis, Miguel A.
AU - Allende, Héctor
PY - 2018/1/1
Y1 - 2018/1/1
N2 - Robotic soccer provides an adversarial scenario where collaborative agents have to execute actions by following a hand-coded or a learned strategy, which in the case of the Small Size League, is given by a centralized decision maker. This work takes advantage of this centralized approach for modelling the keepaway strategy learning problem which is inherently multi-agent, as a single-agent problem, where now each robot forms part of the state of the model. One of the classical reinforcement learning methods is compared with its batch version in terms of amount of time for learning and concluding about updates efficiency based on experiences reusability.
AB - Robotic soccer provides an adversarial scenario where collaborative agents have to execute actions by following a hand-coded or a learned strategy, which in the case of the Small Size League, is given by a centralized decision maker. This work takes advantage of this centralized approach for modelling the keepaway strategy learning problem which is inherently multi-agent, as a single-agent problem, where now each robot forms part of the state of the model. One of the classical reinforcement learning methods is compared with its batch version in terms of amount of time for learning and concluding about updates efficiency based on experiences reusability.
UR - http://www.scopus.com/inward/record.url?scp=85062661949&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85062661949
T3 - CEUR Workshop Proceedings
BT - CRoNe 2018 - Proceedings of the 4th Congress on Robotics and Neuroscience
A2 - Nettle, Cristobal J.
A2 - Solis, Miguel A.
PB - CEUR-WS
T2 - 4th Congress on Robotics and Neuroscience, CRoNe 2018
Y2 - 15 November 2018 through 17 November 2018
ER -