This thesis aims to contribute to the area of kernel methods, which are a class of machine learning methods known for their wide applicability and state-of-the-art performance, but which suffer from high training and evaluation complexity. The work in this thesis utilizes the notion of reduced-set models to alleviate the training and testing complexities of these methods in a unified manner. In the first part of the thesis, we use recent results in kernel smoothing and integral-operator learning to design a generic strategy to speed up various kernel methods. In Chapter 3, we present a method to speed up kernel PCA (KPCA), which is one of the fundamental kernel methods for manifold learning, by using reduced-set density estimates (RSDE) of the data. The proposed method induces an integral operator that is an approximation of the ideal integral operator associated to KPCA. It is shown that the error between the ideal and approximate integral operators is related to the error between the ideal and approximate kernel density estimates of the data. In Chapter 4, we derive similar approximation algorithms for Gaussian process regression, diffusion maps, and kernel embeddings of conditional distributions. In the second part of the thesis, we use reduced-set models for kernel methods to tackle online learning in model-reference adaptive control (MRAC). In Chapter 5, we relate the properties of the feature spaces induced by Mercer kernels to make a connection between persistency-of-excitation and the budgeted placement of kernels to minimize tracking and modeling error. In Chapter 6, we use a Gaussian process (GP) formulation of the modeling error to accommodate a larger class of errors, and design a reduced-set algorithm to learn a GP model of the modeling error. Proofs of stability for all the algorithms are presented, and simulation results on a challenging control problem validate the methods.

Submitted in fulfillment of the requirements of the degree of Doctor of Engineering: Civil Engineering, Durban University of Technology. Durban. South Africa, 2015. / The planning and management of water resources systems often involve formulation and establishment of optimal operating policies and the study of trade-off between different objectives. Due to the intricate nature of water resources management tasks, several models with varying degrees of complexities have been developed and applied for resolving water resources optimisation and allocation problems. Nevertheless, there still exist uncertainties about finding a generally consistent and trustworthy method that can find solutions which are very close to the global optimum in all scenarios. This study presents the development and application of a new evolutionary multi-objective optimisation algorithm, combined Pareto multi-objective differential evolution (CPMDE). The algorithm combines methods of Pareto ranking and Pareto dominance selections to implement a novel generational selection scheme. The new scheme provides a systematic approach for controlling elitism of the population which results in the simultaneous creation of short solution vectors that are suitable for local search and long vectors suitable for global search. By incorporating combined Pareto procedures, CPMDE is able to adaptively balance exploitation of non-dominated solutions found with exploration of the search space. Thus, it is able to escape all local optima and converge to the global Pareto-optimal front. The performance of CPMDE was compared with 14 state-of-the-art evolutionary multi-objective optimisation algorithms. A total of ten test problems and three real world problems were considered in the benchmark of the algorithm. Findings suggest that the new algorithm presents an improvement in convergence to global Pareto-optimal fronts especially on deceptive multi-modal functions where CPMDE clearly outperformed all other algorithms in convergence and diversity. The convergence metric on this problem was several orders of magnitude better than those of the other algorithms. Competitive results obtained from the benchmark of CPMDE suggest that it is a good alternative for solving real multi-objective optimisation problems. Also, values of a variance statistics further indicate that CPMDE is reliable and stable in finding solutions and converging to Pareto-optimal fronts in multi-objective optimisation problems. CPMDE was applied to resolve water allocation problems in the Orange River catchment in South Africa. Results obtained from the applications of CPMDE suggest it represents an improvement over some existing methods. CPMDE was applied to resolve water allocation problems in the agricultural and power sectors in South Africa. These sectors are strategic in forging economic growth, sustaining technological developments and contributing further to the overall development of the nation. They are also germane in capacitating the South African government’s commitment towards equity and poverty eradication and ensuring food security. Harnessing more hydropower from existing water sources within the frontier of the country is germane in capacitating the South African Government’s commitment to reduction of the countries’ greenhouse gas emissions and transition to a low-carbon economy while meeting a national target of 3 725 megawatts by 2030. Application of CPMDE algorithm in the behavioural analysis of the Vanderkloof reservoir showed an increase of 20 310 MWH in energy generation corresponding to a 3.2 percent increase. On analysis of storage trajectories over the operating period, it was found that the real time analysis incorporating a hybrid between CPMDE and ANN offers a procedure with a high ability to minimize deviation from target storage under the prevailing water stress condition. Overall, the real time analysis provides an improvement of 49.32 percent over the current practice. Further analysis involving starting the simulation with a proposed higher storage volume suggests that 728.53 GWH of annual energy may be generated from the reservoir under medium flow condition without system failure as opposed to 629 GWH produced from current practice. This corresponds to a 13.66 percent increase in energy generation. It was however noted that the water resources of the dam is not in excess. The water in the dam is just enough to meet all current demands. This calls for proper management policies for future operation of the reservoir to guard against excessive storage depletions. The study herein also involved the development of a decision support system for the daily operation of the Vanderkloof reservoir. This provides a low cost solution methodology suitable for the sustainable operation of the Vanderkloof dam in South Africa. Adopting real time optimisation strategies may be beneficial to the operation of reservoirs. Findings from the study herein indicate that the new algorithm represents an improvement over existing methods. Therefore, CPMDE presents a new tool that nations can adapt for the proper management of water resources towards the overall prosperity of their populace. / D

A dissertation submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in partial fulfillment of the requirements for the degree of Master of Science. Pretoria 1991. / Remote sensing can be used to produce maps of land-cover, but to be of use to the GIS community these maps must first be vectorized in an intelligent manner. Existing algorithms suffer from the defects of being slow, memory intensive and producing vast quantities of very short vectors. Furthermore if these vectors are thinned via standard algorithms, errors are introduced. The process of vectorizing raster maps is subject to major ambiguities. Thus an infinite family of vector maps ccrresponds to each raster map. This dissertation presents an algorithm for converting raster maps in a rapid manner to accurate vector maps with a minimum of vectors. The algorithm converts raster maps to vector maps using local information only, (a two by two neighbourhood). the method is "exact" in the sense that rasterizing the resulting polygons would produce exactly the same raster map, pixel for pixel. The method is "near optimal" in that it produces, in a local sense, that "exacb" vector map having the least number of vectors. The program is built around a home-grown object oriented Programming System (OOPS) for the C programming language. The main features of the OOPS system, (called OopCdaisy), are virtual and static methods, polymorphism, generalized containers, container indices and thorough error checking, The following general purpose objects are implemented with a large number of sophistiated methods :- Stacks, LIFO lists, scannable containers with indices, trees and 2D objects like points, lines etc. / AC2017

Implementing Shamir's secret sharing scheme using floating point arithmetic would provide a faster and more efficient secret sharing scheme due to the speed in which GPUs perform floating point arithmetic. However, with the loss of a finite field, properties of a perfect secret sharing scheme are not immediately attainable. The goal is to analyze the plausibility of Shamir's secret sharing scheme using floating point arithmetic achieving the properties of a perfect secret sharing scheme and propose improvements to attain these properties. Experiments indicate that property 2 of a perfect secret sharing scheme, "Any k-1 or fewer participants obtain no information regarding the shared secret", is compromised when Shamir's secret sharing scheme is implemented with floating point arithmetic. These experimental results also provide information regarding possible solutions and adjustments. One of which being, selecting randomly generated points from a smaller interval in one of the proposed schemes of this thesis. Further experimental results indicate improvement using the scheme outlined. Possible attacks are run to test the desirable properties of the different schemes and reinforce the improvements observed in prior experiments. / by Timothy Finemore. / Thesis (M.S.)--Florida Atlantic University, 2012. / Includes bibliography. / Electronic reproduction. Boca Raton, Fla., 2012. Mode of access: World Wide Web.

In today's networked world, information is widely distributed across many independent databases in heterogeneous formats. Integrating such information is a difficult task and has been adressed by several projects. However, previous integration solutions, such as the EVE-Project, have several shortcomings. Database contents and structure change frequently, and users often have incomplete information about the data content and structure of the databases they use. When information from several such insufficiently described sources is to be extracted and integrated, two problems have to be solved: How can we discover the structure and contents of and interrelationships among unknown databases, and how can we provide durable integration views over several such databases? In this dissertation, we have developed solutions for those key problems in information integration. The first part of the dissertation addresses the fact that knowledge about the interrelationships between databases is essential for any attempt at solving the information integration problem. We are presenting an algorithm called FIND2 based on the clique-finding problem in graphs and k-uniform hypergraphs to discover redundancy relationships between two relations. Furthermore, the algorithm is enhanced by heuristics that significantly reduce the search space when necessary. Extensive experimental studies on the algorithm both with and without heuristics illustrate its effectiveness on a variety of real-world data sets. The second part of the dissertation addresses the durable view problem and presents the first algorithm for incremental view maintenance in schema-restructuring views. Such views are essential for the integration of heterogeneous databases. They are typically defined in schema-restructuring query languages like SchemaSQL, which can transform schema into data and vice versa, making traditional view maintenance based on differential queries impossible. Based on an existing algebra for SchemaSQL, we present an update propagation algorithm that propagates updates along the query algebra tree and prove its correctness. We also propose optimizations on our algorithm and present experimental results showing its benefits over view recomputation.

In this paper, new software and hardware designs for the NTRU Public Key Cryptosystem are proposed. The first design attempts to improve NTRU's polynomial multiplication through applying techniques from the Chinese Remainder Theorem (CRT) to the convolution algorithm. Although the application of CRT shows promise for the creation of the inverse polynomials in the setup procedure, it does not provide any benefits to the procedures that are critical to the performance of NTRU (public key creation, encryption, and decryption). This research has identified that this is due to the small coefficients of one of the operands, which can be a common misunderstanding. The second design focuses on improving the performance of the polynomial multiplications within NTRU's key creation, encryption, and decryption procedures through hardware. This design exploits the inherent parallelism within a polynomial multiplication to make scalability possible. The advantage scalability provides is that it allows the user to customize the design for low and high power applications. In addition, the support for arbitrary precision allows the user to meet the desired security level. The third design utilizes the Montgomery Multiplication algorithm to develop an unified architecture that can perform a modular multiplication for GF(p) and GF(2^k) and a polynomial multiplication for NTRU. The unified design only requires an additional 10 gates in order for the Montgomery Multiplier core to compute the polynomial multiplication for NTRU. However, this added support for NTRU presents some restrictions on the supported lengths of the moduli and on the chosen value for the residue for the GF(p) and GF(2^k) cases. Despite these restrictions, this unified architecture is now capable of supporting public key operations for the majority of Public-Key Cryptosystems.

Lu Yang. / Thesis submitted in: November 2004. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (leaves 96-103). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgement --- p.iv / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Motivation --- p.1 / Chapter 1.2 --- Our Contributions --- p.3 / Chapter 1.3 --- Thesis Outline --- p.5 / Chapter 2 --- Background Study --- p.7 / Chapter 2.1 --- MPEG-7 Introduction --- p.8 / Chapter 2.2 --- Image Analysis in CBIR Systems --- p.11 / Chapter 2.2.1 --- Color Information --- p.13 / Chapter 2.2.2 --- Color Layout --- p.19 / Chapter 2.2.3 --- Texture Information --- p.20 / Chapter 2.2.4 --- Shape Information --- p.24 / Chapter 2.2.5 --- CBIR Systems --- p.26 / Chapter 2.3 --- Image Processing in JPEG Frequency Domain --- p.30 / Chapter 2.4 --- Photo Album Clustering --- p.33 / Chapter 3 --- Feature Extraction and Similarity Analysis --- p.38 / Chapter 3.1 --- Feature Set in Frequency Domain --- p.38 / Chapter 3.1.1 --- JPEG Frequency Data --- p.39 / Chapter 3.1.2 --- Our Feature Set --- p.42 / Chapter 3.2 --- Digital Photo Similarity Analysis --- p.43 / Chapter 3.2.1 --- Energy Histogram --- p.43 / Chapter 3.2.2 --- Photo Distance --- p.45 / Chapter 4 --- 1-Dimensional Photo Album Management Techniques --- p.49 / Chapter 4.1 --- Photo Album Sorting --- p.50 / Chapter 4.2 --- Photo Album Clustering --- p.52 / Chapter 4.3 --- Photo Album Compression --- p.56 / Chapter 4.3.1 --- Variable IBP frames --- p.56 / Chapter 4.3.2 --- Adaptive Search Window --- p.57 / Chapter 4.3.3 --- Compression Flow --- p.59 / Chapter 4.4 --- Experiments and Performance Evaluations --- p.60 / Chapter 5 --- High Dimensional Photo Clustering --- p.67 / Chapter 5.1 --- Traditional Clustering Techniques --- p.67 / Chapter 5.1.1 --- Hierarchical Clustering --- p.68 / Chapter 5.1.2 --- Traditional K-means --- p.71 / Chapter 5.2 --- Multidimensional Scaling --- p.74 / Chapter 5.2.1 --- Introduction --- p.75 / Chapter 5.2.2 --- Classical Scaling --- p.77 / Chapter 5.3 --- Our Interactive MDS-based Clustering --- p.80 / Chapter 5.3.1 --- Principal Coordinates from MDS --- p.81 / Chapter 5.3.2 --- Clustering Scheme --- p.82 / Chapter 5.3.3 --- Layout Scheme --- p.84 / Chapter 5.4 --- Experiments and Results --- p.87 / Chapter 6 --- Conclusions --- p.94 / Bibliography --- p.96

In recent years, there has been an exponential increase in container volume shipment within intermodal transportation systems. Container terminals as part of the global port system represent important hubs within this intermodal transportation system. Thus, the need to improve the operational efficiency is the most important issue for container terminals from an economic standpoint. Moreover, intermodal transportation systems, ports and inland transport facilities should all be integrated into one coordinated plan. More specifically, a method to schedule different types of handling equipment in an integrated way within a container terminal is a popular topic for researchers. However, not many researchers have addresses this topic in relationship to the simulation aspect which will test feasible solutions under real container terminal environment parameters. In order to increase the efficiency of operations, the development of mathematical models and algorithms is critical in finding the best feasible solution. The objective of this study is to evaluate the feasible solution to find the proper number of Yard Trailers (YTs) with the minimal cost for the container terminals. This study uses the Dynamic YTs operation's method as a background for modeling. A mathematical model with various constraints related to the integrated operations among the different types of handling equipment is formulated. This model takes into consideration both serving time of quay cranes and yard cranes, and cost reduction strategies by decreasing use of YTs with the specific objective of minimum total cost including utilization of YTs and vessel berthing. In addition, a heuristic algorithm combined with Monte Carlo Method and Brute-Force Search are employed. The early Stage Technique of Monte Carlo method is proposed to generate vast random numbers to replicate simulation for real cases. / The Brute-Force Search is used for identifying all potential cases specific to the conditions of this study. Some preliminary numerical test results suggest that this method is good for use in conjunction with simulation of container terminal operation. The expected outcome of this research is a solution to obtain the proper number of YTs for transporting containers with a minimum cost; thus, improving the operational efficiency in a container terminal. / by Yueqiong Zhao. / Thesis (M.S.C.S.)--Florida Atlantic University, 2011. / Includes bibliography. / Electronic reproduction. Boca Raton, Fla., 2011. Mode of access: World Wide Web.

by Johnson, Tin-Chak Pang. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1997. / Includes bibliographical references. / Acknowledgments / Abstract / List of figures / List of tables / Chapter 1. --- Introduction --- p.1-1 / Chapter 1.1 --- Introduction --- p.1-1 / Chapter 1.2 --- Introduction to asynchronous system --- p.1-5 / Chapter 1.2.1 --- Motivation --- p.1-5 / Chapter 1.2.2 --- Hazards --- p.1-7 / Chapter 1.2.3 --- Classes of Asynchronous circuits --- p.1-8 / Chapter 1.3 --- Introduction to Transform Coding --- p.1-9 / Chapter 1.4 --- Organization of the Thesis --- p.1-16 / Chapter 2. --- Asynchronous Design Methodologies --- p.2-1 / Chapter 2.1 --- Introduction --- p.2-1 / Chapter 2.2 --- Self-timed system --- p.2-2 / Chapter 2.3 --- DCVSL Methodology --- p.2-4 / Chapter 2.3.1 --- DCVSL gate --- p.2-5 / Chapter 2.3.2 --- Handshake Control --- p.2-7 / Chapter 2.4 --- Micropipeline Methodology --- p.2-11 / Chapter 2.4.1 --- Summary of previous design --- p.2-12 / Chapter 2.4.2 --- New Micropipeline structure and improvements --- p.2-17 / Chapter 2.4.2.1 --- Asymmetrical delay --- p.2-20 / Chapter 2.4.2.2 --- Variable Delay and Delay Value Selection --- p.2-22 / Chapter 2.5 --- Comparison between DCVSL and Micropipeline --- p.2-25 / Chapter 3. --- Self-timed Multipliers --- p.3-1 / Chapter 3.1 --- Introduction --- p.3-1 / Chapter 3.2 --- Design Example 1 : Bit-serial matrix multiplier --- p.3-3 / Chapter 3.2.1 --- DCVSL design --- p.3-4 / Chapter 3.2.2 --- Micropipeline design --- p.3-4 / Chapter 3.2.3 --- The first test chip --- p.3-5 / Chapter 3.2.4 --- Second test chip --- p.3-7 / Chapter 3.3 --- Design Example 2 - Modified Booth's Multiplier --- p.3-9 / Chapter 3.3.1 --- Circuit Design --- p.3-10 / Chapter 3.3.2 --- Simulation result --- p.3-12 / Chapter 3.3.3 --- The third test chip --- p.3-14 / Chapter 4. --- Current-Sensing Completion Detection --- p.4-1 / Chapter 4.1 --- Introduction --- p.4-1 / Chapter 4.2 --- Current-sensor --- p.4-2 / Chapter 4.2.1 --- Constant current source --- p.4-2 / Chapter 4.2.2 --- Current mirror --- p.4-4 / Chapter 4.2.3 --- Current comparator --- p.4-5 / Chapter 4.3 --- Self-timed logic using CSCD --- p.4-9 / Chapter 4.4 --- CSCD test chips and testing results --- p.4-10 / Chapter 4.4.1 --- Test result --- p.4-11 / Chapter 5. --- Self-timed ICT processor architecture --- p.5-1 / Chapter 5.1 --- Introduction --- p.5-1 / Chapter 5.2 --- Comparison of different architecture --- p.5-3 / Chapter 5.2.1 --- General purpose Digital Signal Processor --- p.5-5 / Chapter 5.2.1.1 --- Hardware and speed estimation : --- p.5-6 / Chapter 5.2.2 --- Micropipeline without fast algorithm --- p.5-7 / Chapter 5.2.2.1 --- Hardware and speed estimation : --- p.5-8 / Chapter 5.2.3 --- Micropipeline with fast algorithm (I) --- p.5-8 / Chapter 5.2.3.1 --- Hardware and speed estimation : --- p.5-9 / Chapter 5.2.4 --- Micropipeline with fast algorithm (II) --- p.5-10 / Chapter 5.2.4.1 --- Hardware and speed estimation : --- p.5-11 / Chapter 6. --- Implementation of self-timed ICT processor --- p.6-1 / Chapter 6.1 --- Introduction --- p.6-1 / Chapter 6.2 --- Implementation of Self-timed 2-D ICT processor (First version) --- p.6-3 / Chapter 6.2.1 --- 1-D ICT module --- p.6-4 / Chapter 6.2.2 --- Self-timed Transpose memory --- p.6-5 / Chapter 6.2.3 --- Layout Design --- p.6-8 / Chapter 6.3 --- Implementation of Self-timed 1-D ICT processor with fast algorithm (final version) --- p.6-9 / Chapter 6.3.1 --- I/O buffers and control units --- p.6-10 / Chapter 6.3.1.1 --- Input control --- p.6-11 / Chapter 6.3.1.2 --- Output control --- p.6-12 / Chapter 6.3.1.2.1 --- Self-timed Computational Block --- p.6-13 / Chapter 6.3.1.3 --- Handshake Control Unit --- p.6-14 / Chapter 6.3.1.4 --- Integer Execution Unit (IEU) --- p.6-18 / Chapter 6.3.1.5 --- Program memory and Instruction decoder --- p.6-20 / Chapter 6.3.2 --- Layout Design --- p.6-21 / Chapter 6.4 --- Specifications of the final version self-timed ICT chip --- p.6-22 / Chapter 7. --- Testing of Self-timed ICT processor --- p.7-1 / Chapter 7.1 --- Introduction --- p.7-1 / Chapter 7.2 --- Pin assignment of Self-timed 1 -D ICT chip --- p.7-2 / Chapter 7.3 --- Simulation --- p.7-3 / Chapter 7.4 --- Testing of Self-timed 1-D ICT processor --- p.7-5 / Chapter 7.4.1 --- Functional test --- p.7-5 / Chapter 7.4.1.1 --- Testing environment and results --- p.7-5 / Chapter 7.4.2 --- Transient Characteristics --- p.7-7 / Chapter 7.4.3 --- Comments on speed and power --- p.7-10 / Chapter 7.4.4 --- Determination of optimum delay control voltage --- p.7-12 / Chapter 7.5 --- Testing of delay element and other logic cells --- p.7-13 / Chapter 8. --- Conclusions --- p.8-1 / Bibliography / Appendices

Wong, Ka Chun. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2010. / Includes bibliographical references (leaves 109-121). / Abstracts in English and Chinese. / Abstract --- p.ii / Acknowledgements --- p.iv / List of Figures --- p.ix / List of Tables --- p.xi / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Motivation --- p.1 / Chapter 1.2 --- Objective --- p.2 / Chapter 1.3 --- Methodology --- p.2 / Chapter 1.4 --- Bioinforrnatics --- p.2 / Chapter 1.5 --- Computational Methods --- p.3 / Chapter 1.5.1 --- Evolutionary Algorithms --- p.3 / Chapter 1.5.2 --- Data Mining for TF-TFBS bindings --- p.4 / Chapter 2 --- Background --- p.5 / Chapter 2.1 --- Gene Transcription --- p.5 / Chapter 2.1.1 --- Protein-DNA Binding --- p.6 / Chapter 2.1.2 --- Existing Methods --- p.6 / Chapter 2.1.3 --- Related Databases --- p.8 / Chapter 2.1.3.1 --- TRANSFAC - Experimentally Determined Database --- p.8 / Chapter 2.1.3.2 --- cisRED - Computational Determined Database --- p.9 / Chapter 2.1.3.3 --- ORegAnno - Community Driven Database --- p.10 / Chapter 2.2 --- Evolutionary Algorithms --- p.13 / Chapter 2.2.1 --- Representation --- p.15 / Chapter 2.2.2 --- Parent Selection --- p.16 / Chapter 2.2.3 --- Crossover Operators --- p.17 / Chapter 2.2.4 --- Mutation Operators --- p.18 / Chapter 2.2.5 --- Survival Selection --- p.19 / Chapter 2.2.6 --- Termination Condition --- p.19 / Chapter 2.2.7 --- Discussion --- p.19 / Chapter 2.2.8 --- Examples --- p.19 / Chapter 2.2.8.1 --- Genetic Algorithm --- p.20 / Chapter 2.2.8.2 --- Genetic Programming --- p.21 / Chapter 2.2.8.3 --- Differential Evolution --- p.21 / Chapter 2.2.8.4 --- Evolution Strategy --- p.22 / Chapter 2.2.8.5 --- Swarm Intelligence --- p.23 / Chapter 2.3 --- Association Rule Mining --- p.24 / Chapter 2.3.1 --- Objective --- p.24 / Chapter 2.3.2 --- Apriori Algorithm --- p.24 / Chapter 2.3.3 --- Partition Algorithm --- p.25 / Chapter 2.3.4 --- DHP --- p.25 / Chapter 2.3.5 --- Sampling --- p.25 / Chapter 2.3.6 --- Frequent Pattern Tree --- p.26 / Chapter 3 --- Discovering Protein-DNA Binding Sequence Patterns Using Associa- tion Rule Mining --- p.27 / Chapter 3.1 --- Materials and Methods --- p.28 / Chapter 3.1.1 --- Association Rule Mining and Apriori Algorithm --- p.29 / Chapter 3.1.2 --- Discovering associated TF-TFBS sequence patterns --- p.29 / Chapter 3.1.3 --- "Data, Preparation" --- p.31 / Chapter 3.2 --- Results and Analysis --- p.34 / Chapter 3.2.1 --- Rules Discovered --- p.34 / Chapter 3.2.2 --- Quantitative Analysis --- p.36 / Chapter 3.2.3 --- Annotation Analysis --- p.37 / Chapter 3.2.4 --- Empirical Analysis --- p.37 / Chapter 3.2.5 --- Experimental Analysis --- p.38 / Chapter 3.3 --- Verifications --- p.41 / Chapter 3.3.1 --- Verification by PDB --- p.41 / Chapter 3.3.2 --- Verification by Homology Modeling --- p.45 / Chapter 3.3.3 --- Verification by Random Analysis --- p.45 / Chapter 3.4 --- Discussion --- p.49 / Chapter 4 --- Designing Evolutionary Algorithms for Multimodal Optimization --- p.50 / Chapter 4.1 --- Introduction --- p.50 / Chapter 4.2 --- Problem Definition --- p.51 / Chapter 4.2.1 --- Minimization --- p.51 / Chapter 4.2.2 --- Maximization --- p.51 / Chapter 4.3 --- An Evolutionary Algorithm with Species-specific Explosion for Multi- modal Optimization --- p.52 / Chapter 4.3.1 --- Background --- p.52 / Chapter 4.3.1.1 --- Species Conserving Genetic Algorithm --- p.52 / Chapter 4.3.2 --- Evolutionary Algorithm with Species-specific Explosion --- p.53 / Chapter 4.3.2.1 --- Species Identification --- p.53 / Chapter 4.3.2.2 --- Species Seed Delta Evaluation --- p.55 / Chapter 4.3.2.3 --- Stage Switching Condition --- p.56 / Chapter 4.3.2.4 --- Species-specific Explosion --- p.57 / Chapter 4.3.2.5 --- Calculate Explosion Weights --- p.59 / Chapter 4.3.3 --- Experiments --- p.59 / Chapter 4.3.3.1 --- Performance measurement --- p.60 / Chapter 4.3.3.2 --- Parameter settings --- p.61 / Chapter 4.3.3.3 --- Results --- p.61 / Chapter 4.3.4 --- Conclusion --- p.62 / Chapter 4.4 --- A. Crowding Genetic. Algorithm with Spatial Locality for Multimodal Op- timization --- p.64 / Chapter 4.4.1 --- Background --- p.64 / Chapter 4.4.1.1 --- Crowding Genetic Algorithm --- p.64 / Chapter 4.4.1.2 --- Locality of Reference --- p.64 / Chapter 4.4.2 --- Crowding Genetic Algorithm with Spatial Locality --- p.65 / Chapter 4.4.2.1 --- Motivation --- p.65 / Chapter 4.4.2.2 --- Offspring generation with spatial locality --- p.65 / Chapter 4.4.3 --- Experiments --- p.67 / Chapter 4.4.3.1 --- Performance measurements --- p.67 / Chapter 4.4.3.2 --- Parameter setting --- p.68 / Chapter 4.4.3.3 --- Results --- p.68 / Chapter 4.4.4 --- Conclusion --- p.68 / Chapter 5 --- Generalizing Protein-DNA Binding Sequence Representations and Learn- ing using an Evolutionary Algorithm for Multimodal Optimization --- p.70 / Chapter 5.1 --- Introduction and Background --- p.70 / Chapter 5.2 --- Problem Definition --- p.72 / Chapter 5.3 --- Crowding Genetic Algorithm with Spatial Locality --- p.72 / Chapter 5.3.1 --- Representation --- p.72 / Chapter 5.3.2 --- Crossover Operators --- p.73 / Chapter 5.3.3 --- Mutation Operators --- p.73 / Chapter 5.3.4 --- Fitness Function --- p.74 / Chapter 5.3.5 --- Distance Metric --- p.76 / Chapter 5.4 --- Experiments --- p.77 / Chapter 5.4.1 --- Parameter Setting --- p.77 / Chapter 5.4.2 --- Search Space Estimation --- p.78 / Chapter 5.4.3 --- Experimental Procedure --- p.78 / Chapter 5.4.4 --- Results and Analysis --- p.79 / Chapter 5.4.4.1 --- Generalization Analysis --- p.79 / Chapter 5.4.4.2 --- Verification By PDB --- p.86 / Chapter 5.5 --- Conclusion --- p.87 / Chapter 6 --- Predicting Protein Structures on a Lattice Model using an Evolution- ary Algorithm for Multimodal Optimization --- p.88 / Chapter 6.1 --- Introduction --- p.88 / Chapter 6.2 --- Problem Definition --- p.89 / Chapter 6.3 --- Representation --- p.90 / Chapter 6.4 --- Related Works --- p.91 / Chapter 6.5 --- Crowding Genetic Algorithm with Spatial Locality --- p.92 / Chapter 6.5.1 --- Motivation --- p.92 / Chapter 6.5.2 --- Customization --- p.92 / Chapter 6.5.2.1 --- Distance metrics --- p.92 / Chapter 6.5.2.2 --- Handling infeasible conformations --- p.93 / Chapter 6.6 --- Experiments --- p.94 / Chapter 6.6.1 --- Performance Metrics --- p.94 / Chapter 6.6.2 --- Parameter Settings --- p.94 / Chapter 6.6.3 --- Results --- p.94 / Chapter 6.7 --- Conclusion --- p.95 / Chapter 7 --- Conclusion and Future Work --- p.97 / Chapter 7.1 --- Thesis Contribution --- p.97 / Chapter 7.2 --- Fixture Work --- p.98 / Chapter A --- Appendix --- p.99 / Chapter A.1 --- Problem Definition in Chapter 3 --- p.107 / Bibliography --- p.109 / Author's Publications --- p.122