Genetic Programming + Multi-Agent Reinforcement Learning: Hybrid Approaches for Decision Processes

Created by W.Langdon from gp-bibliography.bib Revision:1.8051

@InProceedings{Fitch:2022:AERO,

• author = "Natalie Fitch and Daniel Clancy",
• title = "Genetic Programming + Multi-Agent Reinforcement Learning: Hybrid Approaches for Decision Processes",
• booktitle = "2022 IEEE Aerospace Conference (AERO)",
• year = "2022",
• month = "5-12 " # mar,
• address = "Big Sky, MT, USA",
• keywords = "genetic algorithms, genetic programming, Training, Q-learning, Sensitivity, Heuristic algorithms, Atmospheric modeling, Games",
• ISSN = "1095-323X",
• isbn13 = "978-1-6654-3761-5",
• DOI = "doi:10.1109/AERO53065.2022.9843637",
• abstract = "This paper details progress within the Multi-Agent Reinforcement Learning (MARL) research area with application to agent decision processing in complex battle-space scenarios, including air, surface, sub-surface, and space domains. We implement a Double Deep Q-Network (DDQN) with Minimax Q-Learning in order to model simultaneous, zero-sum, two team engagements involving multiple Blue agents & Red opponents. This is a game theoretic approach that models both ally and opponent policies while viewing a battle as a Multi-Stage Markov Stochastic Game (MSMSG). We contrast our agent with a DDQN + Traditional Q-Learning algorithm in a single stage 2v1 battle scenario with mixed optimal strategies. In order to help mitigate learning sensitivities and local optima convergence, we implement a Genetic Programming (GP) algorithm, which outperforms both the Minimax Q-Learning and Traditional Q-Learning DDQN agents trained using traditional stochastic gradient descent in a dynamic 1v1 battle. Lastly, we create a hybrid approach that combines stochastic gradient descent learning (Minimax Q-Learning) and gradient-free learning (GP) and apply our hybrid approach within the StarCraft II (SC2) 3m map, which simulates a 3v3 battle. We contrast this hybrid MARL approach with another state-of-the-art MARL method (QMIX) for the SC2 3m combat scenario.",
• notes = "Also known as \cite{9843637}",

}

Genetic Programming entries for Natalie Fitch Daniel Clancy

Citations