Automatic Structure Generation using Genetic Programming and Fractal Geometry

Three dimensional model design is a well-known and studied field, with numerous real-world applications. However, the manual construction of these models can often be time-consuming to the average user, despite the advantages o ffered through computational advances. This thesis presents an approach to the design of 3D structures using evolutionary computation and L-systems, which involves the automated production of such designs using a strict set of fitness functions. These functions focus on the geometric properties of the models produced, as well as their quantifiable aesthetic value - a topic which has not been widely investigated with respect to 3D models. New extensions to existing aesthetic measures are discussed and implemented in the presented system in order to produce designs which are visually pleasing. The system itself facilitates the construction of models requiring minimal user initialization and no user-based feedback throughout the evolutionary cycle. The genetic programming evolved models are shown to satisfy multiple criteria, conveying a relationship between their assigned aesthetic value and their perceived aesthetic value. Exploration into the applicability and e ffectiveness of a multi-objective approach to the problem is also presented, with a focus on both performance and visual results. Although subjective, these results o er insight into future applications and study in the fi eld of computational aesthetics and automated structure design.

Focus Group Report

Report to Terry O'Malley from Tom highlighting favourable responses from a focus group on the "Butter It" campaign.

Attention focus and balance control in young and older adults

This thesis investigated attention focus and balance control in eighteen healthy young adults and eighteen healthy older adults. All participants performed sixteen consecutive trials of a balance task which involved standing for 30-s on an unstable platform that could rotate only in the roll direction. There were no attention focus instructions provided on any of the sixteen trials. Following the completion of the initial and final attempt in the series, participants reported "where" their attention had been focused when performing the task. The results showed differences in balance between young and older adults and improvements in balance with practice in both young and older adults. However, there were no differences in attention focus strategies between young and older adults. Both age groups directed attention to multiple sources during the balance task. An equal focus on internal (i.e., feet, trunk, and other body parts) and external (i.e., the platform) sources with little focus on events not related to the task dominated on the first attempt of the balance task. Focus on internal sources was maintained and focus on events not related to the task increased at the expense of focus on external sources on the final attempt of the balance task. Following the series of sixteen trials to establish "natural" attention focus, participants performed three randomly presented trials, each with specific attention focus instructions (i.e., think about minimizing movements of the feet, the trunk, or the platform). The results showed that, in contrast to the literature, instructions to focus on an internal source, the trunk, actually augmented control of the task as reflected in reduced trunk sway whereas instructions to focus on an internal source, the feet, or an external source, the platform, did not benefit performance on the task. Thus, the distance fi-om the interaction point of the body with the external source is critical and may not depend on whether the source is internal or external. Thus, a global attention focus instruction may not be beneficial and the nature of the task should be considered when adopting attention focus instructions for young and older adults.

Automatic evolution of conceptual building architectures

This thesis describes research in which genetic programming is used to automatically evolve shape grammars that construct three dimensional models of possible external building architectures. A completely automated fitness function is used, which evaluates the three dimensional building models according to different geometric properties such as surface normals, height, building footprint, and more. In order to evaluate the buildings on the different criteria, a multi-objective fitness function is used. The results obtained from the automated system were successful in satisfying the multiple objective criteria as well as creating interesting and unique designs that a human-aided system might not discover. In this study of evolutionary design, the architectures created are not meant to be fully functional and structurally sound blueprints for constructing a building, but are meant to be inspirational ideas for possible architectural designs. The evolved models are applicable for today's architectural industries as well as in the video game and movie industries. Many new avenues for future work have also been discovered and highlighted.

Automatic Inference of Graph Models for Complex Networks with Genetic Programming

Complex networks can arise naturally and spontaneously from all things that act as a part of a larger system. From the patterns of socialization between people to the way biological systems organize themselves, complex networks are ubiquitous, but are currently poorly understood. A number of algorithms, designed by humans, have been proposed to describe the organizational behaviour of real-world networks. Consequently, breakthroughs in genetics, medicine, epidemiology, neuroscience, telecommunications and the social sciences have recently resulted. The algorithms, called graph models, represent significant human effort. Deriving accurate graph models is non-trivial, time-intensive, challenging and may only yield useful results for very specific phenomena. An automated approach can greatly reduce the human effort required and if effective, provide a valuable tool for understanding the large decentralized systems of interrelated things around us. To the best of the author's knowledge this thesis proposes the first method for the automatic inference of graph models for complex networks with varied properties, with and without community structure. Furthermore, to the best of the author's knowledge it is the first application of genetic programming for the automatic inference of graph models. The system and methodology was tested against benchmark data, and was shown to be capable of reproducing close approximations to well-known algorithms designed by humans. Furthermore, when used to infer a model for real biological data the resulting model was more representative than models currently used in the literature.

The effects of consistency and inconsistency between attentional focus and task objective in learning a golf putting task

Converging evidence has demonstrated learning advantages when an individual is instructed to focus their attention externally. However, many of the motor tasks utilized in past research had clear external objectives (i.e., putting accuracy), creating a compatible relationship between an external focus of attention (i.e., outcome) and an external task objective (i.e., putting accuracy). The present study examined whether or not the consistency of instructions and task objective would differentially impact the acquisition of a golf putting task. Participants performed a putting task in a control condition or in one of four experimental conditions resulting from the factorial interaction of task instructions (internal or external) and task objective (internal or external). The retention and transfer data revealed that participants who received an external task objective demonstrated superior outcome scores. Participants who received technique information paired with outcome information demonstrated superior technique scores.

HUMAN GENOME VARIATIONS AND EVOLUTION WITH A FOCUS ON THE ANALYSIS OF TRANSPOSABLE ELEMENTS

Genome sequence varies in numerous ways among individuals although the gross architecture is fixed for all humans. Retrotransposons create one of the most abundant structural variants in the human genome and are divided in many families, with certain members in some families, e.g., L1, Alu, SVA, and HERV-K, remaining active for transposition. Along with other types of genomic variants, retrotransponson-derived variants contribute to the whole spectrum of genome variants in humans. With the advancement of sequencing techniques, many human genomes are being sequenced at the individual level, fueling the comparative research on these variants among individuals. In this thesis, the evolution and functional impact of structural variations is examined primarily focusing on retrotransposons in the context of human evolution. The thesis comprises of three different studies on the topics that are presented in three data chapters. First, the recent evolution of all human specific AluYb members, representing the second most active subfamily of Alus, was tracked to identify their source/master copy using a novel approach. All human-specific AluYb elements from the reference genome were extracted, aligned with one another to construct clusters of similar copies and each cluster was analyzed to generate the evolutionary relationship between the members of the cluster. The approach resulted in identification of one major driver copy of all human specific Yb8 and the source copy of the Yb9 lineage. Three new subfamilies within the AluYb family – Yb8a1, Yb10 and Yb11 were also identified, with Yb11 being the youngest and most polymorphic. Second, an attempt to construct a relation between transposable elements (TEs) and tandem repeats (TRs) was made at a genome-wide scale for the first time. Upon sequence comparison, positional cross-checking and other relevant analyses, it was observed that over 20% of all TRs are derived from TEs. This result established the first connection between these two types of repetitive elements, and extends our appreciation for the impact of TEs on genomes. Furthermore, only 6% of these TE-derived TRs follow the already postulated initiation and expansion mechanisms, suggesting that the others are likely to follow a yet-unidentified mechanism. Third, by taking a combination of multiple computational approaches involving all types of genetic variations published so far including transposable elements, the first whole genome sequence of the most recent common ancestor of all modern human populations that diverged into different populations around 125,000-100,000 years ago was constructed. The study shows that the current reference genome sequence is 8.89 million base pairs larger than our common ancestor’s genome, contributed by a whole spectrum of genetic mechanisms. The use of this ancestral reference genome to facilitate the analysis of personal genomes was demonstrated using an example genome and more insightful recent evolutionary analyses involving the Neanderthal genome. The three data chapters presented in this thesis conclude that the tandem repeats and transposable elements are not two entirely distinctly isolated elements as over 20% TRs are actually derived from TEs. Certain subfamilies of TEs themselves are still evolving with the generation of newer subfamilies. The evolutionary analyses of all TEs along with other genomic variants helped to construct the genome sequence of the most recent common ancestor to all modern human populations which provides a better alternative to human reference genome and can be a useful resource for the study of personal genomics, population genetics, human and primate evolution.

Real-Time Automatic Object Classification and Tracking using Genetic Programming and NVIDIA R CUDA TM

Genetic Programming (GP) is a widely used methodology for solving various computational problems. GP's problem solving ability is usually hindered by its long execution times. In this thesis, GP is applied toward real-time computer vision. In particular, object classification and tracking using a parallel GP system is discussed. First, a study of suitable GP languages for object classification is presented. Two main GP approaches for visual pattern classification, namely the block-classifiers and the pixel-classifiers, were studied. Results showed that the pixel-classifiers generally performed better. Using these results, a suitable language was selected for the real-time implementation. Synthetic video data was used in the experiments. The goal of the experiments was to evolve a unique classifier for each texture pattern that existed in the video. The experiments revealed that the system was capable of correctly tracking the textures in the video. The performance of the system was on-par with real-time requirements.

Network Similarity Measures and Automatic Construction of Graph Models using Genetic Programming

A complex network is an abstract representation of an intricate system of interrelated elements where the patterns of connection hold significant meaning. One particular complex network is a social network whereby the vertices represent people and edges denote their daily interactions. Understanding social network dynamics can be vital to the mitigation of disease spread as these networks model the interactions, and thus avenues of spread, between individuals. To better understand complex networks, algorithms which generate graphs exhibiting observed properties of real-world networks, known as graph models, are often constructed. While various efforts to aid with the construction of graph models have been proposed using statistical and probabilistic methods, genetic programming (GP) has only recently been considered. However, determining that a graph model of a complex network accurately describes the target network(s) is not a trivial task as the graph models are often stochastic in nature and the notion of similarity is dependent upon the expected behavior of the network. This thesis examines a number of well-known network properties to determine which measures best allowed networks generated by different graph models, and thus the models themselves, to be distinguished. A proposed meta-analysis procedure was used to demonstrate how these network measures interact when used together as classifiers to determine network, and thus model, (dis)similarity. The analytical results form the basis of the fitness evaluation for a GP system used to automatically construct graph models for complex networks. The GP-based automatic inference system was used to reproduce existing, well-known graph models as well as a real-world network. Results indicated that the automatically inferred models exemplified functional similarity when compared to their respective target networks. This approach also showed promise when used to infer a model for a mammalian brain network.

Can Attention Focus Instructions Reduce the Effects of Fatigue on Balance Control?

Localized muscular fatigue has been identified to have detrimental effects on balance control, an important skill for everyday life. Manipulation of attention focus instructions has been shown to benefit performance of various motor skills including balance and has been found to facilitate endurance during fatiguing tasks. The purpose of this thesis was to determine if the use of attention focus instructions could attenuate the effects of muscular fatigue on balance control. Twenty-four participants performed a balance task (two-legged stance on an unstable platform) before and after a fatigue protocol. Trunk sway, platform excursions, and lower limb muscle activity was measured. Results suggest that use of either internal or external attention focus instructions can reduce the immediate effects of muscular fatigue of the lower limb on balance control as shown through reduced trunk sway and platform excursions. These results have relevance for individuals performing balance tasks in a fatigued state.

The Influence of Attentional Focus on the Self-Efficacy-Performance Relationship in a Continuous Running Task

The self-efficacy-performance relationship in continuous sport tasks has been shown to be significantly reciprocal yet unequal with stronger influences in the performance-to-self-efficacy pathway rather than self-efficacy-to-performance pathway (e.g., LaForge-MacKenzie & Sullivan, 2014b). Bandura (2012) suggested that sociocognitive variables may influence this relationship. Attention as a sociocognitve factor may bias the processing of performance and self-efficacy information (Bandura, 1982, 1997; Bandura & Jourden, 1991). As confidence and attention are important aspects of peak running performance (Brewer, Van Raalte, Linder, & VanRaalte, 1991), the primary purpose of the present study was to examine the self-efficacy-performance relationship under three conditions of attentional focus. The secondary purpose was to examine self-efficacy and performance as separate constructs under the same conditions of attention. Participants ran continuously for one kilometer in one of three randomly assigned attentional focus conditions: internal-focus (n = 51), external-focus (n = 50), and control (n = 49). Self-efficacy was assessed using a one-item measure every 200 meters. Path analyses examining the primary objective revealed significant self-efficacy-to-performance pathways in all conditions: external-focus (p < .05, βs ranging from -.17 to -.32), internal-focus (p < .05, βs ranging from -.26 to -.36), and control (p < .05, βs ranging from -.29 to -.42). Significant reciprocal relationships were absent in all conditions. ANOVAs examining the secondary objectives found significantly faster performance in the control condition at the start (F (2, 147) = 3.86, p < .05) and end of the task (F (2, 147) = 3.56, p < .05). Self-efficacy was significantly higher in the internal-focus condition at the end of the task (Self-Efficacy 4 (F (2, 147) = 3.21, p < .05) and Self-Efficacy 5 (F (2, 147) = 4.74, p < .05). In contrast to previous within-trial research (e.g., LaForge-MacKenzie & Sullivan, 2014b) self-efficacy-to-performance effects were more significant and robust than performance-to-self-efficacy effects. These results provided support for Bandura’s (2012) suggestion that sociocognitive factors may have the ability to alter the causal structure of the self-efficacy-performance relationship, proposing complexities in the self-efficacy-performance relationship (Sitzmann &Yeo, 2013). Results were discussed from both theoretical and applied perspectives.

Automatic Preparation of ETD Material from the Internet Archive for the DSpace Repository Platform

A big challenge associated with getting an institutional repository off the ground is getting content into it. This article will look at how to use digitization services at the Internet Archive alongside software utilities that the author developed to automate the harvesting of scanned dissertations and associated Dublin Core XML files to create an ETD Portal using the DSpace platform. The end result is a metadata-rich, full-text collection of theses that can be constructed for little out of pocket cost.

Object-Oriented Genetic Programming for the Automatic Inference of Graph Models for Complex Networks

Complex networks are systems of entities that are interconnected through meaningful relationships. The result of the relations between entities forms a structure that has a statistical complexity that is not formed by random chance. In the study of complex networks, many graph models have been proposed to model the behaviours observed. However, constructing graph models manually is tedious and problematic. Many of the models proposed in the literature have been cited as having inaccuracies with respect to the complex networks they represent. However, recently, an approach that automates the inference of graph models was proposed by Bailey [10] The proposed methodology employs genetic programming (GP) to produce graph models that approximate various properties of an exemplary graph of a targeted complex network. However, there is a great deal already known about complex networks, in general, and often specific knowledge is held about the network being modelled. The knowledge, albeit incomplete, is important in constructing a graph model. However it is difficult to incorporate such knowledge using existing GP techniques. Thus, this thesis proposes a novel GP system which can incorporate incomplete expert knowledge that assists in the evolution of a graph model. Inspired by existing graph models, an abstract graph model was developed to serve as an embryo for inferring graph models of some complex networks. The GP system and abstract model were used to reproduce well-known graph models. The results indicated that the system was able to evolve models that produced networks that had structural similarities to the networks generated by the respective target models.

The Impact of Directed Mirror Focus and Technique Cues on Psychological, Cognitive and Emotional Exercise Correlates in an Introductory Weight Training Orientation

The purpose of the study was to investigate whether teaching inactive and low active women to use mirrors for form and technique purposes could lessen the negative impact of mirrors on self-presentational concerns, affect, and self-efficacy. Eligible women (N = 82) underwent a one-on-one weight training orientation with a personal trainer. Participants were randomized into one of four experimental groups, each unique in the type of feedback (general or technique-specific) and the degree of focus on the mirror for technique reinforcement. Questionnaires assessed study outcomes pre- and post-orientation. Results indicated groups did not significantly differ on any post-condition variables, when controlling for pre-condition values (all p’s >.05). All groups showed outcome improvements following the orientation. This suggests that during a complex task, a personal trainer who emphasizes form and technique can facilitate improvements to psychological outcomes in novice exercisers, independent of the presence of mirrors or directional cues provided.

Drivers of Local Firms’ Focus on Internal R&D in Emerging Economies and Their International Performance

Grounded on the resource-based view of the firm, the study of this thesis investigates the effect of four internal and external factors – engineer intensity, location, affiliation with the government, government funding – on Chinese firms’ decision to either invest in internal R&D activities or external R&D and the effect of this decision on the firms’ international market success. In addition, the moderating role of the presence of foreign firms in China is examined. To understand these relationships, the thesis’ theorization focuses on the issue of how firms can combine optimally the two options – “internal R&D” and “external R&D”. In this regard I juxtapose internal R&D and external R&D and compare their advantages and disadvantages. To test my model, I apply panel data from the Annual Industrial Survey Database provided by the Chinese National Bureau of Statistics. My results show that three of the four investigated factors affect Chinese firms’ resource allocation decisions; and effective resource allocation decisions lead effectively to international market success, strengthened by the presence of foreign firms in China. Moreover the findings bear several theoretical and managerial contributions. First I propose the last dimension of the “VRIO framework” – “organization” – as an endogenous component of the VRIO framework, as my study investigated how firms can effectively combine resources to generate a competitive advantage in terms of international market success. Previous academic literature so far focused on examining whether internal and external R&D are complements or substitutes. My study fills a gap in the literature by investigating the determinants of the efficient combination of the two strategies and the outcome of the combination. One of the managerial implications is that Chinese firms can learn from foreign companies that are present in China.