Document Type
Honors Thesis
Abstract
In Monte-Carlo simulation, we often need an estimate of the domain of the target objective function of the sample. We want to constrain the domain within a specific shape to sample more points from it, efficiently. That is why it is good to bind the domain of the objective function that we are trying to sample from. This research aims to develop an unsupervised clustering algorithm that works on a multidimensional dataset and gives considerable output using far less time than other similar algorithms, decreasing computational expenses by a significant amount. The algorithm gives us the predicted number of ellipsoids, their covariance matrices, and centers. Benchmarking of the algorithm uses the Dynesty package for Python. There is a discussion of the benefits and shortcomings of the algorithm and a discussion of future steps to make the algorithm more universal at the end.
Publication Date
5-1-2021
Language
English
License
This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License.
Recommended Citation
Chapagain, Nabin, "COMPUTATIONAL GEOMETRY INSIGHTS INTO MONTE-CARLO SIMULATION METHODS" (2021). 2021 Spring Honors Capstone Projects. 29.
https://mavmatrix.uta.edu/honors_spring2021/29