Here are a few of the wargaming articles that have come across my desk recently…
C. Lázaro Carrascosa, I. P. Ylardia, M. Paredes-Velasco and M. d. C. N. García-Suelto, “Game-Based Learning With Augmented Reality for History Education,” in IEEE Revista Iberoamericana de Tecnologias del Aprendizaje, vol. 19, pp. 14-23, 2024, doi: 10.1109/RITA.2024.3368348. Keywords: {History; Education; Games; Augmented reality; Visualization; Videos; Technological innovation; Palabras clave- ICT; augmented reality; game-based learning (GBL); innovation in history learning; learning difficulties}
Abstract: One of the fundamental aspects of history teaching is the chronology of important events and facts that have occurred. However, this involves an effort of memorization that produces rejection in students, affecting their learning outcomes. This article describes an experience with Primary and Secondary students using Augmented Reality with gamification in timeline activities to learn a course of History. The results show that there was improvement in learning in several aspects: chronologically situate several events in relation to BC and AD, relate different historical events to ages in History, and sort important historical milestones. In addition, it was found that students experienced positive emotions during the experience such as joy, curiosity, and interest.
My key reads:
- “In fact, there are already findings indicating that students consider the use of immersive technology to study History to be highly suitable, particularly in visualizing historical events.”
- “Lastly, other researchers approach [gamification] from the perspective of game thinking, where ‘gamification is presenting any kind of process as if it were a game,’ and players ‘should have fun while achieving the objectives of the gamified process.’ In essence, gamification aims to achieve specific goals and/or provide rewarding learning experiences.”
- “The findings indicate that the use of AR with gamification methodologies improves learning outcomes and has a positive impact on student motivation and emotion.”
J. Coble, A. Barton, C. Darken and S. Black, “Optimizing Naval Movement Using Deep Reinforcement Learning,” 2023 International Conference on Machine Learning and Applications (ICMLA), Jacksonville, FL, USA, 2023, pp. 400-407, doi: 10.1109/ICMLA58977.2023.00062. Keywords: {Decision making; Games; Deep reinforcement learning; Real-time systems; Marine vehicles; artificial intelligence; machine learning; reinforcement learning; MCTS; wargaming}
Abstract: In a rapidly evolving maritime warfare landscape, the U.S. Navy and its allies require their crews to quickly identify optimal strategies for vessel engagements to ensure freedom of the seas. This necessity becomes more pronounced given the potential grave consequences of sub-optimal maneuvers, as illustrated by the cases of the USS John McCain and USS Fitzgerald. Recent advancements in Machine Learning (ML) and Artificial Intelligence (AI) offer a promising solution. There have been significant strides in implementing AI to outperform human experts in complex games such as Chess, Poker, and StarCraft that now have the potential to also benefit real-time decision-making and wargaming in the naval domain. This study explores the potential for Reinforcement Learning (RL) techniques to be applied in naval contexts, which could provide valuable decision-support tools to ship captains and their staffs by suggesting optimal movement strategies in complex maritime scenarios. In this study, exemplar naval scenarios were designed and modeled within a combat simulation environment, AI agents (consisting of a mix of rule-based, method-based, and value-based approaches) were designed, and the performances of these agents were evaluated and compared. The aim was to assess the agents’ ability to identify optimal movements against a rule-based adversary, while also comparing these performances against human-level play. The insights drawn from this study contribute to ongoing research aimed at developing effective decision aids for ship captains in real-world operations.
My key reads:
- “Because wargames usually involve two competing sides, these simulated conflicts provide an opportunity to leverage intelligent agent behavior development due to the zero-sum relationship between the sides (i.e., one side’s victory is the other side’s loss), which provides more distinct feedback in the learning process.”
- “A significant finding is the exceptional performance of the DQN [Deep Q-Networks] agent over all other AI agents assessed in this study. The DQN agent demonstrated a promising ability to identify optimal strategies in different situations—outperforming human players in some of the scenarios presented. DQN’s robustness and adaptive nature allowed it to generalize and adapt to different operational contexts, thus making it an asset to investigate further application in the U.S. Navy’s decision-making process.”
- “Our results indicate that RL-based AI agents, with additional refinement and customization for naval context, could offer substantial advantages for both training simulations and real-world naval operations. Although more work needs to be done in this domain, the promising outcomes of the study suggest a potential role for AI in providing robust support to decision-makers in the U.S. Navy and beyond.”
Lamparth, M., Corso, A., Ganz, J., Mastro, O.S., Schneider, J., & Trinkunas, H. (2024) Human vs. Machine: Language Models and Wargames. arXiv:2403.03407. https://doi.org/10.48550/arXiv.2403.03407.
Abstract: Wargames have a long history in the development of military strategy and the response of nations to threats or attacks. The advent of artificial intelligence (AI) promises better decision-making and increased military effectiveness. However, there is still debate about how AI systems, especially large language models (LLMs), behave as compared to humans. To this end, we use a wargame experiment with 107 national security expert human players designed to look at crisis escalation in a fictional US-China scenario and compare human players to LLM-simulated responses. We find considerable agreement in the LLM and human responses but also significant quantitative and qualitative differences between simulated and human players in the wargame, motivating caution to policymakers before handing over autonomy or following AI-based strategy recommendations.
My key reads:
- “The experimental results show that the simulated wargames with the LLMs have largely consistent responses to the scenario and treatments with human players.”
- “We identified a significant overlap in general behavior between human players and those simulated by LLMs. However, we found notable differences in the strategic preferences between humans and the tested LLMs. The LLM simulations demonstrated sensitivity to command instructions or simulating the dialog, no sensitivity to player background attributes, and engaged in farcical discussions among players. These results offer a unique comparison between expert human players and LLMs that complements the recent work, which shows that, if left to their own devices and when acting independently, LLMs can lead to arms-race dynamics and show escalatory tendencies.”
- “Without rigorous testing, detailed deployment criteria, and new technical approaches enabling behavioral guarantees, decision-makers should be cautious about using LLMs as direct substitutes for human recommendations.”
Feature image courtesy forum.canardpc.com
The opinions and views expressed in this blog are those of the author alone and are presented in a personal capacity. They do not necessarily represent the views of U.S. Navy or any other U.S. government Department, Agency, Office, or employer.
RockyMountainNavy.com © 2007-2024 by Ian B is licensed under CC BY-SA 4.0 
Rocky Mountain Navy Gamer 