You are here
Search results
(1 - 2 of 2)
- Title
- A differential evolution approach to feature selection in genomic prediction
- Creator
- Whalen, Ian
- Date
- 2018
- Collection
- Electronic Theses & Dissertations
- Description
-
The use of genetic markers has become widespread for prediction of genetic merit in agricultural applications and is a beginning to show promise for estimating propensity to disease in human medicine. This process is known as genomic prediction and attempts to model the mapping between an organism's genotype and phenotype. In practice, this process presents a challenging problem. Sequencing and recording phenotypic traits are often expensive and time consuming. This leads to datasets often...
Show moreThe use of genetic markers has become widespread for prediction of genetic merit in agricultural applications and is a beginning to show promise for estimating propensity to disease in human medicine. This process is known as genomic prediction and attempts to model the mapping between an organism's genotype and phenotype. In practice, this process presents a challenging problem. Sequencing and recording phenotypic traits are often expensive and time consuming. This leads to datasets often having many more features than samples. Common models for genomic prediction often fall victim to overfitting due to the curse of dimensionality. In this domain, only a fraction of the markers that are present significantly affect a particular trait. Models that fit to non-informative markers are in effect fitting to statistical noise, leading to a decrease in predictive performance. Therefore, feature selection is desirable to remove markers that do not appear to have a significant effect on the trait being predicted. The method presented here uses differential evolution based search for feature selection. This study will characterize differential evolution's efficacy in feature selection for genomic prediction and present several extensions to the base search algorithm in an attempt to apply domain knowledge to guide the search toward better solutions.
Show less
- Title
- EMERGENT COORDINATION : ADAPTATION, OPEN-ENDEDNESS, AND COLLECTIVE INTELLIGENCE
- Creator
- Bao, Honglin
- Date
- 2022
- Collection
- Electronic Theses & Dissertations
- Description
-
Agent-based modeling is a widely used computational method for studying the micro-macro bridge issue by simulating the microscopic interactions and observing the macroscopic emergence. This thesis begins with the fundamental methodology of agent-based models: how agents are represented, how agents interact, and how the agent population is structured. Two vital topics, the evolution of cooperation and opinion dynamics are used to illustrate methodological innovation. For the first topic, we...
Show moreAgent-based modeling is a widely used computational method for studying the micro-macro bridge issue by simulating the microscopic interactions and observing the macroscopic emergence. This thesis begins with the fundamental methodology of agent-based models: how agents are represented, how agents interact, and how the agent population is structured. Two vital topics, the evolution of cooperation and opinion dynamics are used to illustrate methodological innovation. For the first topic, we study the equilibrium selection in a coordination game in multi-agent systems. In particular, we focus on the characteristics of agents (supervisors and subordinates versus representative agents), the interactions of agents (reinforcement learning in the games with fixed versus adaptive learning rates according to the supervision and time-varying versus supervision-guided exploration rates), the network of agents (single-layer versus multi-layer networks), and their impact on the emergent behaviors. Regarding the second topic, we examine how opinions evolve and spread in a cognitively heterogeneous agent population with sparse interactions and how the opinion dynamics co-evolve with the open-ended society's structural change. We then discuss the rich insights into collective intelligence in the two proposed models viewed from the interaction-based adaptation and open-ended network structure. We finally link collective emergent intelligence to diverse applications in the realm of computing and other scientific fields in a cross-multidisciplinary manner.
Show less