Distributed Architecture

Simulation-as-a-Service for Reinforcement Learning Applications by Example of Heavy Plate Rolling Processes
In this paper we propose a distributed architecture which allows the remote training of reinforcement learning agents on a simulation.
Christian Scheiderer, Timo Thun, Christian Idzik, Andres Felipe Posada-Moreno, Alexander Krämer, Johannes Lohmar, Gerhard Hirt, Tobias Meisen
Cite Project DOI URL