Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 4 of 4
Full-Text Articles in Computer Engineering
Generation Of Strategies For Environmental Deception In Two-Player Normal-Form Games, Howard E. Poston
Generation Of Strategies For Environmental Deception In Two-Player Normal-Form Games, Howard E. Poston
Theses and Dissertations
Methods of performing and defending against deceptive actions are a popular field of study in game theory; however, the focus is mostly on action deception in turn-based games. This work focuses on developing strategies for performing environmental deception in two-player, strategic-form games. Environmental deception is defined as deception where one player has the ability to change the other's perception of the state of the game through modification of their perception of the game's payoff matrix, similar to the use of camouflage. The main contributions of this research are an expansion of the definition of the stability of a Nash equilibrium …
Counter Weapon Control, Brian J. Roadruck
Counter Weapon Control, Brian J. Roadruck
Theses and Dissertations
In this work, pursuit-evasion differential game theory is applied to the target defense scenario. The targets, attackers, and the defenders positional information is assumed to be known to the players. With positional information, a computational efficient method of strategy synthesis can be derived applying a differential game theory approach. We demonstrate that when non-optimal strategies are employed by one of the players, e.g. Line Of Sight guidance, the outcome will favor the players that employ the optimal strategy given by the solution of the pursuit-evasion differential game.
Robust Models For Operator Workload Estimation, Andrew M. Smith
Robust Models For Operator Workload Estimation, Andrew M. Smith
Theses and Dissertations
When human-machine system operators are overwhelmed, judicious employment of automation can be beneficial. Ideally, a system which can accurately estimate current operator workload can make better choices when to employ automation. Supervised machine learning models can be trained to estimate workload in real time from operator physiological data. Unfortunately, estimating operator workload using trained models is limited: using a model trained in one context can yield poor estimation of workload in another. This research examines the utility of three algorithms (linear regression, regression trees, and Artificial Neural Networks) in terms of cross-application workload prediction. The study is conducted for a …
Leveraging Human Insights By Combining Multi-Objective Optimization With Interactive Evolution, Joshua R. Christman
Leveraging Human Insights By Combining Multi-Objective Optimization With Interactive Evolution, Joshua R. Christman
Theses and Dissertations
Deceptive fitness landscapes are a growing concern for evolutionary computation. Recent work has shown that combining human insights with short-term evolution has a synergistic effect that accelerates the discovery of solutions. While humans provide rich insights, they fatigue easily. Previous work reduced the number of human evaluations by evolving a diverse set of candidates via intermittent searches for novelty. While successful at evolving solutions for a deceptive maze domain, this approach lacks the ability to measure what the human evaluator identifies as important. The key insight here is that multi-objective evolutionary algorithms foster diversity, serving as a surrogate for novelty, …