USING NEURAL NETWORKS TO DETERMINE COURSE OF ACTION FOR A LAND-BASED CONSTRUCTIVE SIMULATION

Simple item page
dc.contributor.advisorDarken, Christian J.
dc.contributor.authorLian, Weiwen Mervyn
dc.contributor.departmentComputer Science (CS)
dc.contributor.secondreaderAlt, Jonathan K.
dc.date.accessioned2019-11-04T18:20:00Z
dc.date.available2019-11-04T18:20:00Z
dc.date.issued2019-09
dc.description.abstractThe latest breakthroughs achieved by various commercial companies in applying neural networks to games have led to research in applying such technologies to the military domain. This thesis uses a proxy stochastic simulation environment to identify the factors required to train a neural network via reinforcement learning to maneuver forces in a battlefield and overcome hostile forces. An incremental training approach was used to train the neural network to recognize the enemy first, before training it to consolidate forces as a form of emergent behavior. Eight neural networks were trained successfully to maneuver forces to engage the enemies. Seven of these neural networks managed to generalize their training to search for enemies in a larger area than the scenario in which they were trained. The neural networks also showed some success in performing force consolidation before attacking an enemy. The most important neural network hyperparameters that contributed to training success were training duration, learning rate, discount factor, and loss factor. This thesis also found that the neural network reward function and randomizing starting positions during the training phase are critical to training success.en_US
dc.description.distributionstatementApproved for public release; distribution is unlimited.
dc.description.serviceCivilian, Ministry of Defense, Singaporeen_US
dc.description.urihttp://archive.org/details/usingneuralnetwo1094563476
dc.identifier.thesisid32728
dc.identifier.urihttps://hdl.handle.net/10945/63476
dc.publisherMonterey, CA; Naval Postgraduate Schoolen_US
dc.rightsCopyright is reserved by the copyright owner.en_US
dc.subject.authorneural networken_US
dc.subject.authorcourse of actionen_US
dc.titleUSING NEURAL NETWORKS TO DETERMINE COURSE OF ACTION FOR A LAND-BASED CONSTRUCTIVE SIMULATIONen_US
dc.typeThesisen_US
dspace.entity.typePublication
etd.thesisdegree.disciplineModeling, Virtual Environments, and Simulation Institute (MOVES)en_US
etd.thesisdegree.grantorNaval Postgraduate Schoolen_US
etd.thesisdegree.levelMastersen_US
etd.thesisdegree.nameMaster of Science in Modeling, Virtual Environments, and Simulationen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
19Sep_Lian_Weiwen Mervyn.pdf
Size:
2.07 MB
Format:
Adobe Portable Document Format
Download
Collections
Theses