Fok Ka Ling. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (leaves 96-101). / Abstracts in English and Chinese. / Abstract --- p.1 / Acknowledgement --- p.iv / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Evolutionary Computation --- p.1 / Chapter 1.2 --- Graphics Processing Unit --- p.2 / Chapter 1.3 --- Objective --- p.3 / Chapter 1.4 --- Contribution --- p.4 / Chapter 1.5 --- Thesis Organization --- p.4 / Chapter 2 --- Evolutionary Computation --- p.6 / Chapter 2.1 --- Introduction --- p.6 / Chapter 2.2 --- General Framework --- p.7 / Chapter 2.3 --- Features of Evolutionary Algorithm --- p.8 / Chapter 2.3.1 --- Widely Applicable --- p.8 / Chapter 2.3.2 --- Parallelism --- p.9 / Chapter 2.3.3 --- Robust to Change --- p.9 / Chapter 2.4 --- Parallel and Distributed Evolutionary Algorithm --- p.9 / Chapter 2.4.1 --- Global Parallel Evolutionary Algorithms --- p.10 / Chapter 2.4.2 --- Fine-Grained Evolutionary Algorithms --- p.11 / Chapter 2.4.3 --- Island Distributed Evolutionary Algorithms --- p.12 / Chapter 2.5 --- Summary --- p.14 / Chapter 3 --- Graphics Processing Unit --- p.15 / Chapter 3.1 --- Introduction --- p.15 / Chapter 3.2 --- History of GPU --- p.16 / Chapter 3.2.1 --- First-Generation GPUs --- p.16 / Chapter 3.2.2 --- Second-Generation GPUs --- p.17 / Chapter 3.2.3 --- Third-Generation GPUs --- p.17 / Chapter 3.2.4 --- Fourth-Generation GPUs --- p.17 / Chapter 3.3 --- The Graphics Pipelining --- p.18 / Chapter 3.3.1 --- Standard Graphics Pipeline --- p.18 / Chapter 3.3.2 --- Programmable Graphics Pipeline --- p.18 / Chapter 3.3.3 --- Fragment Processors for Scientific Computation --- p.21 / Chapter 3.4 --- GPU-CPU Analogy --- p.23 / Chapter 3.4.1 --- Memory Architecture --- p.23 / Chapter 3.4.2 --- Processing Model --- p.24 / Chapter 3.5 --- Limitation of GPU --- p.24 / Chapter 3.5.1 --- Limited Input and Output --- p.24 / Chapter 3.5.2 --- Slow Data Readback --- p.24 / Chapter 3.5.3 --- No Random Number Generator --- p.25 / Chapter 3.6 --- Summary --- p.25 / Chapter 4 --- Evolutionary Programming on GPU --- p.26 / Chapter 4.1 --- Introduction --- p.26 / Chapter 4.2 --- Evolutionary Programming --- p.26 / Chapter 4.3 --- Data Organization --- p.29 / Chapter 4.4 --- Fitness Evaluation --- p.31 / Chapter 4.4.1 --- Introduction --- p.31 / Chapter 4.4.2 --- Different Forms of Fitness Function --- p.32 / Chapter 4.4.3 --- Parallel Fitness Function Evaluation using GPU --- p.33 / Chapter 4.5 --- Mutation --- p.34 / Chapter 4.5.1 --- Introduction --- p.34 / Chapter 4.5.2 --- Self Adaptive Mutation Operators --- p.36 / Chapter 4.5.3 --- Mutation on GPU --- p.37 / Chapter 4.6 --- Selection for Replacement --- p.39 / Chapter 4.6.1 --- Introduction --- p.39 / Chapter 4.6.2 --- Classification of Selection Operator --- p.39 / Chapter 4.6.3 --- q -Tournament Selection --- p.40 / Chapter 4.6.4 --- Median Searching --- p.41 / Chapter 4.6.5 --- Minimizing Data Transfer --- p.43 / Chapter 4.7 --- Experimental Results --- p.44 / Chapter 4.7.1 --- Visualization --- p.48 / Chapter 4.8 --- Summary --- p.49 / Chapter 5 --- Genetic Algorithm on GPU --- p.56 / Chapter 5.1 --- Introduction --- p.56 / Chapter 5.2 --- Canonical Genetic Algorithm --- p.57 / Chapter 5.2.1 --- Parent Selection --- p.57 / Chapter 5.2.2 --- Crossover and Mutation --- p.62 / Chapter 5.2.3 --- Replacement --- p.63 / Chapter 5.3 --- Experiment Results --- p.64 / Chapter 5.4 --- Summary --- p.66 / Chapter 6 --- Multi-Objective Genetic Algorithm --- p.70 / Chapter 6.1 --- Introduction --- p.70 / Chapter 6.2 --- Definitions --- p.71 / Chapter 6.2.1 --- General MOP --- p.71 / Chapter 6.2.2 --- Decision Variables --- p.71 / Chapter 6.2.3 --- Constraints --- p.71 / Chapter 6.2.4 --- Feasible Region --- p.72 / Chapter 6.2.5 --- Optimal Solution --- p.72 / Chapter 6.2.6 --- Pareto Optimum --- p.73 / Chapter 6.2.7 --- Pareto Front --- p.73 / Chapter 6.3 --- Multi-Objective Genetic Algorithm --- p.75 / Chapter 6.3.1 --- Ranking --- p.76 / Chapter 6.3.2 --- Fitness Scaling --- p.77 / Chapter 6.3.3 --- Diversity Preservation --- p.77 / Chapter 6.4 --- A Niched and Elitism Multi-Objective Genetic Algorithm on GPU --- p.79 / Chapter 6.4.1 --- Objective Values Evaluation --- p.80 / Chapter 6.4.2 --- Pairwise Pareto Dominance and Pairwise Distance --- p.81 / Chapter 6.4.3 --- Fitness Assignment --- p.85 / Chapter 6.4.4 --- Embedded Archiving Replacement --- p.87 / Chapter 6.5 --- Experiment Result --- p.89 / Chapter 6.6 --- Summary --- p.90 / Chapter 7 --- Conclusion --- p.95 / Bibliography --- p.96

Previous models of language acquisition have had partial success describing the processes that children use to acquire knowledge of the grammatical categories of their native language. The present study used a computer model based on the evolutionary principles of adaptation and selection to gain further insight into children's acquisition of grammatical categories. Transcribed language samples of eight parents or caregivers each conversing with their own child served as the input corpora for the model. The model was tested on each child's language corpus three times: two fixed mutation rates as well as a progressively decreasing mutation rate, which allowed less adaptation over time, were examined. The output data were evaluated by measuring the computer model's ability to correctly identify the grammatical categories in 500 utterances from the language corpus of each child. The model's performance ranged between 78 and 88 percent correct; the highest performance overall was found for a corpus using the progressively decreasing mutation rate, but overall no clear pattern relative to mutation rate was found.

Advanced technologies, a world-wide trend to deregulation of power systems and environmental constraints have attracted increasing interest in the operation of electric power systems with multiple storage resources. Under the competitive pressure of the deregulation, new efficient solution techniques to adapt quickly to the changing marketplace are in demand. This thesis presents a new evolutionary method, Constructive Evolutionary Programming (CEP), for minimising the system operational cost of scheduling electric power systems with multiple storage resources. The method combines the advantages of Constructive Dynamic Programming and Evolutionary Programming. Instead of evolving the "primal" variables such as storage content releases and thermal generator outputs, CEP evolves the piecewise linear convex cost-to-go functions (i.e. the storage content value curves). The multi-stage problem of multi-storage power system scheduling is thus decomposed into many smaller one-stage subproblems with evolved cost-to-go functions. For each evolutionary individual, linear programming is used in the forward pass process to solve the dispatch subproblems and the total system operational cost over the scheduling period is assigned to its fitness. Case studies demonstrate that the proposed method is robust and efficient for multi-storage power systems, particularly large complex hydrothermal system with cascaded and pumped storages. Although the proposed method is in the early stage of development, relying on assumptions of piecewise linear convexity in a deterministic environment, methods for the incorporation of stochastic models, electrical network and nonlinear, non-convex and non-smooth models are discussed. In addition, a number of possible improvements are also outlined. Due to its simplicity but robustness and efficiency, there are potential research directions for the further development of this evolutionary optimisation method.

Advanced technologies, a world-wide trend to deregulation of power systems and environmental constraints have attracted increasing interest in the operation of electric power systems with multiple storage resources. Under the competitive pressure of the deregulation, new efficient solution techniques to adapt quickly to the changing marketplace are in demand. This thesis presents a new evolutionary method, Constructive Evolutionary Programming (CEP), for minimising the system operational cost of scheduling electric power systems with multiple storage resources. The method combines the advantages of Constructive Dynamic Programming and Evolutionary Programming. Instead of evolving the "primal" variables such as storage content releases and thermal generator outputs, CEP evolves the piecewise linear convex cost-to-go functions (i.e. the storage content value curves). The multi-stage problem of multi-storage power system scheduling is thus decomposed into many smaller one-stage subproblems with evolved cost-to-go functions. For each evolutionary individual, linear programming is used in the forward pass process to solve the dispatch subproblems and the total system operational cost over the scheduling period is assigned to its fitness. Case studies demonstrate that the proposed method is robust and efficient for multi-storage power systems, particularly large complex hydrothermal system with cascaded and pumped storages. Although the proposed method is in the early stage of development, relying on assumptions of piecewise linear convexity in a deterministic environment, methods for the incorporation of stochastic models, electrical network and nonlinear, non-convex and non-smooth models are discussed. In addition, a number of possible improvements are also outlined. Due to its simplicity but robustness and efficiency, there are potential research directions for the further development of this evolutionary optimisation method.

Advanced technologies, a world-wide trend to deregulation of power systems and environmental constraints have attracted increasing interest in the operation of electric power systems with multiple storage resources. Under the competitive pressure of the deregulation, new efficient solution techniques to adapt quickly to the changing marketplace are in demand. This thesis presents a new evolutionary method, Constructive Evolutionary Programming (CEP), for minimising the system operational cost of scheduling electric power systems with multiple storage resources. The method combines the advantages of Constructive Dynamic Programming and Evolutionary Programming. Instead of evolving the "primal" variables such as storage content releases and thermal generator outputs, CEP evolves the piecewise linear convex cost-to-go functions (i.e. the storage content value curves). The multi-stage problem of multi-storage power system scheduling is thus decomposed into many smaller one-stage subproblems with evolved cost-to-go functions. For each evolutionary individual, linear programming is used in the forward pass process to solve the dispatch subproblems and the total system operational cost over the scheduling period is assigned to its fitness. Case studies demonstrate that the proposed method is robust and efficient for multi-storage power systems, particularly large complex hydrothermal system with cascaded and pumped storages. Although the proposed method is in the early stage of development, relying on assumptions of piecewise linear convexity in a deterministic environment, methods for the incorporation of stochastic models, electrical network and nonlinear, non-convex and non-smooth models are discussed. In addition, a number of possible improvements are also outlined. Due to its simplicity but robustness and efficiency, there are potential research directions for the further development of this evolutionary optimisation method.

No description available.

Thesis (M.S.)--Missouri University of Science and Technology, 2008. / Vita. The entire thesis text is included in file. Title from title screen of thesis/dissertation PDF file (viewed April 26, 2008) Includes bibliographical references (p. 65-68).

Thesis (M. E.)--University of New South Wales, 2000. / Also available online.

Thesis (M.S.)--Missouri University of Science and Technology, 2009. / Vita. The entire thesis text is included in file. Title from title screen of thesis/dissertation PDF file (viewed August 10, 2009) Includes bibliographical references (p. 49-51).

The process of acquiring language requires children to learn grammatical categories and apply these categories to new words. Researchers have proposed various explanations of this process in the form of algorithms and computational modeling. Recently, adaptation and selection models have been tested and applied as a possible explanation to the process of acquiring grammatical categories. These studies have proven promising, however, the external validity of this approach has not been examined by grammatically coding samples outside the training corpus. The current thesis applies an adaptation and selection model, which pauses the evolution of dictionaries after every thousand cycles to allow the tagging of 30 outside samples, which are then checked for tagging accuracy. The accuracy across the five training corpora by the six thousandth cycle averaged 76.75%. Additional research is needed to explore the effects of altering the parameters in the model.