High Performance Programming for Soft Computing 1º edição

Marc Notes: Includes bibliographical references and index.; This book examines the present and future of soft computer techniques. It explains how to use the latest technological tools, such as multicore processors and graphics processing units, to implement highly efficient intelligent system methods using a general purpose computer. --; Provided by publisher. Table of Contents: Preface -- 1. Fundamentals of Soft Computing -- 1.1. Introduction -- 1.2. Type-1 Fuzzy Sets -- 1.3. Fuzzy Logic Systems -- 1.4. Type 2 Fuzzy Sets -- 1.5. Type-2 Fuzzy Inference Systems -- 1.5.1. Interval Type-2 fuzzification stage -- 1.5.2. Interval Type-2 inference stage -- 1.5.3. Type-reduction and defuzzification in an interval 13 Type-2 FIS -- 1.6. Artificial Neural Networks -- 1.6.1. A single neuron -- 1.6.2. ANN architectures -- 1.6.3. The learning process -- 1.6.4. The back-propagation algorithm -- 1.7. Adaptive Neuro-Fuzzy Inference System -- 1.7.1. ANFIS -- 1.7.2. AN FIS architecture -- 1.7.3. Hybrid learning algorithm -- References -- 2. Introduction to Novel Microprocessor Architectures -- 2.1. Introduction -- 2.2. History of Computer Processors -- 2.2.1. Beginning of the computer processors -- 2.2.2. Modern commercial multicore processors -- 2.3. Basic Components of a Computer -- 2.4. Characteristics of Modern Processors -- 2.5. Classification of Computer Architectures -- 2.5.1. SISD -- 2.5.2. SIMD -- 2.5.3. MISD -- 2.5.4. MIMD -- 2.6. Parallel Computer Memory Architectures -- 2.7. Measuring Performance -- 2.7.1. Computer performance -- 2.7.2. Analyzing the performance in multicore/multiprocessor systems -- 2.8. Conclusions -- References -- 3. High-Performance Optimization Methods -- 3.1. Introduction -- 3.2. Unconstrained Iterative Optimization Methods -- 3.2.1. Gradient descent method -- 3.2.2. Newton's method -- 3.2.3. Conjugate gradient method -- 3.3. Unconstrained Optimization Using Parallel Computing -- 3.3.1. MATLAB mex-file implementation -- 3.3.2. Testing algorithm performance -- 3.4. Global Optimization Using Memetic Algorithms -- 3.4.1. Set up of the implemented MA -- 3.4.2. Experiments and results -- 3.5. Concluding Remarks -- 3.6. Apendix -- References -- 4. Graphics Processing Unit Programming and Applications -- 4.1. Introduction -- 4.2. CUDA: Compute Unified Device Architecture -- 4.2.1. Programming model -- 4.2.2. Kernel functions -- 4.2.3. Thread hierarchy -- 4.2.4. Memory spaces and hierarchy -- 4.3. CUDA Programming: Fractal Generation and Display -- 4.3.1. Code analysis -- 4.3.2. Benchmarking GPU vs. CPU implementation -- 4.4. Conclusions -- References -- 5. GPU-based Conjugated Gradient Solution for Phase Field Models -- 5.1. Introduction -- 5.2. Conjugate Gradient Method -- 5.2.1. Solution of a linear system as an optimization problem -- 5.2.2. Steepest descent (SD) method -- 5.2.3. Conjugate directions method -- 5.2.4. Linear conjugate gradient method -- 5.2.5. Non-linear conjugate gradient method -- 5.3. Phase Field Model -- 5.3.1. Discretization of the Allen-Cahn equation -- 5.3.2. CUSPARSE and CUBLAS -- 5.3.3. GPU-based solution for the A-C equation -- 5.4. Results -- 5.4.1. 1-Dcase -- 5.4.2. 2-Dcase -- 5.5. Conclusions and Future Work -- References -- 6. Parallel Computing Applied to a Diffusive Model -- 6.1. Introduction -- 6.2. Random Walk -- 6.3. Continuous Distributions -- 6.4. Solution via Finite Difference Approach -- 6.5. High Performance Computing Resources -- 6.6. Conclusions -- References -- 7. Ant Colony Optimization (Past, Present and Future) -- 7.1. Introduction -- 7.1.1. Introduction to swarm intelligence -- 7.1.2. Biological inspiration -- 7.2. The Simple ACO Algorithm -- 7.3. Ant System -- 7.3.1. Tour construction in ant system -- 7.3.2. Update of pheromone trails -- 7.3.3. ACO for the Traveling Salesman Problem -- 7.4. Elitist AS: An Extension of AS -- 7.5. Ant-Q: Introduction to Q-Learning -- 7.6. Ant Colony System (ACS) -- 7.7. MAX-MLAF AS -- 7.8. Rank-based AS -- 7.9. ANTS -- 7.10. BWAS: 2000 -- 7.11. 10 Hyper-Cube AS (HC-AS): 2001 -- 7.12. The Present of ACO: High Performance Computing -- 7.12.1. Introduction to parallel ant colony optimization -- 7.13. Application of SACO in Mobile Robotics -- 7.14. The Future of ACO: A Final Conclusion -- References -- 8. Tool Path Optimization Based on Ant Colony Optimization for CNC Machining Operations -- 8.1. Introduction -- 8.2. Introduction to CNC Programming using CAD/CAM -- 8.2.1. Numerical control programming -- 8.2.2. The use of Computer Aided Manufacturing (CAM) -- 8.2.3. Drilling holes -- 8.2.4. Absolute and incremental programming -- 8.2.5. Canned cycles -- 8.3. Ant Colony Optimization for Hole Making NC Sequences: A Special Case of the Traveling Salesman Problem (TSP) -- 8.3.1. Tour construction in ant system -- 8.3.2. Update of pheromone trails -- 8.3.3. NC for drilling operations using ant colony optimization: Case Study -- 8.4. Parallel Implementation of Ant Colony Optimization -- 8.4.1. Parallel Implementation: Problem Formulation -- 8.4.2. Performance of Parallel Computers -- 8.5. Experimental Results -- 8.5.1. Analysis of the parallel implementation of ACO -- 8.5.2. Tool path optimization analysis -- 8.6. Conclusions -- References -- 9. A Compendium of Artificial Immune Systems -- 9.1. Introduction -- 9.2. A Brief History of Natural and Artificial Immune Systems -- 9.3. Interesting Properties of the Immune System -- 9.4. What Exactly is an Artificial Immune System? -- 9.5. Main Algorithms 202 9.5.1 Negative selection algorithm -- 9.5.2. Clonal selection -- 9.6. Immune Networks -- 9.7. Other Sources of Inspiration -- 9.8. Are Artificial Immune Systems Worth It? -- 9.9. Conclusions -- References -- 10. Applications of Artificial Immune Algorithms -- 10.1. Introduction -- 10.2. Step 1. Reduction -- 10.2.1. City representation using a Class -- 10.2.2. Tour representation using a Class -- 10.2.3. Artificial Vaccines representation using the Vaccine Class -- 10.2.4. Vaccine Generation by Random Selector (VRS) -- 10.2.5. Vaccine Generation by Elitist Selector (VES) -- 10.3. Step 2: Optimization -- 10.4. Step 3: Expansion -- 10.5. Example of Vaccinated TSP -- 10.6. Example: Application of the ROE Method and a GA -- 10.7. Conclusions -- References -- 11. A Parallel Implementation of the NSGA-II -- 11.1. Introduction -- 11.2. Basic Concepts -- 11.3. Genetic Algorithms -- 11.3.1. Crossover operators -- 11.3.2. Mutation operators -- 11.4. NSGA-II -- 11.5. Parallel Implementation of NSGA-II (PNSGA-II) -- 11.6. Results -- 11.7. Conclusions -- References -- 12. High-performance Navigation System for Mobile Robots -- 12.1. Introduction -- 12.2. Artificial Potential Field -- 12.2.1. Attraction potential -- 12.2.2. Repulsive potential -- 12.2.3. Limitations of me artificial potential field -- 12.3. Genetic Algorithms -- 12.3.1. How does a genetic algorithm work? -- 12.3.2. Genetic algorithm operators -- 12.4. High-performance Implementation -- 12.4.1. Phase 1: Simple navigation system with artificial potential field -- 12.4.2. Phase 2: Complete navigation system with artificial potential field and genetic algorithms -- 12.4.3. Phase 3: High-performance navigation system with artificial potential field and parallel genetic algorithms -- 12.5. Results and Conclusions -- References -- 13. A Method Using a Combination of Ant Colony Optimization Variants with Ant Set Partitioning -- 13.1. Introduction -- 13.1.1. Ant Colony Optimization (ACO) -- 13.1.2. ACO variations -- 13.2. Proposed Method -- 13.2.1. Methodology -- 13.3. Experiments -- 13.4. Simulation Results -- 13.4.1. Berlin 52 cities -- 13.4.2. Bier 127 (127 cities) -- 13.5. Conclusions -- References -- 14. Variants of Ant Colony Optimization: A Metaheuristic for Solving the Traveling Salesman Problem -- 14.1. Introduction -- 14.2. ACO Variants -- 14.2.1. Traveling Salesman Problem (TSP) -- 14.2.2. Elitist Ant System -- 14.2.3. Rank based ant system -- 14.2.4. Max-Min ant system -- 14.2.5. Ant Colony System (ACS) -- 14.3. Graphical Interface in Matlab -- 14.4. Sequential Processing -- 14.5. Parallel Processing -- 14.6. Simulation Results -- 14.6.1. Speedup -- 14.7. Conclusions -- References -- 15. Quantum Computing -- 15.1. Introduction -- 15.2. Classic Computation -- 15.3. Basic Mathematics Used in Quantum Computing -- 15.4. Quantum Mechanic: bask Principles -- 15.5. Elements of Quantum Computing -- 15.5.1. The Bloch sphere -- 15.5.2. Quantum registers -- 15.5.3. Quantum measurements -- 15.5.4. Quantum Gates -- 15.5.5. Quantum circuits -- 15.6. Concluding Remarks -- References -- Index -- Color Plate Section. Publisher Marketing: This text examines the present and future of soft computer techniques. It explains how to use the latest technological tools, such as multicore processors and graphics processing units, to implement highly efficient intelligent system methods using a general purpose computer.