Project for a semi-centralized logic-based MARL reward shaping method that is scalable in the number of agents and evaluates it in multiple scenarios
Primary LanguageJupyter NotebookApache License 2.0Apache-2.0