In this work, we propose a new method for obtaining heuristics based on Reinforcement Learning (RL). The synthesis algorithm is thus framed as an RL task with an unbounded action space and a modified version of DQN is used. With a simple and general set of features that abstracts both states and actions, we show that it is possible to learn heuristics on small versions of a problem that generalize to the larger instances, effectively doing zero-shot policy transfer. Our agents learn from scratch in a highly partially observable RL task and outperform the existing heuristic overall, in instances unseen during training. ; Agencia Nacional de Promoción de la Investigación, el Desarrollo Tecnológico y la Innovación ; Universidad de Buenos Aires ; Agencia Nacional de Investigación e Innovación
Exploration Policies for On-the-Fly Controller Synthesis: A Reinforcement Learning Approach
2023-07-08
33rd International Conference on Automated Planning and Scheduling. Prague, Czech Republic. 2023
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
DDC: | 629 |
DEEP REINFORCEMENT LEARNING FOR OPTIMIZING CARPOOLING POLICIES
Europäisches Patentamt | 2019
|DEEP REINFORCEMENT LEARNING FOR OPTIMIZING CARPOOLING POLICIES
Europäisches Patentamt | 2020
|Safe Exploration in Model-Based Reinforcement Learning
Springer Verlag | 2023
|Improve PID controller through reinforcement learning
IEEE | 2018
|