Parallel Genetic Algorithms for Financial Pattern Discovery Using GPUs (eBook)

João Baúto, Rui Neves, Nuno Horta (Autoren)

eBook Download: PDF

2018 | 1st ed. 2018
XIV, 91 Seiten
Springer International Publishing (Verlag)
9783319733296 (ISBN)

Infos

This Brief presents a study of SAX/GA, an algorithm to optimize market trading strategies, to understand how the sequential implementation of SAX/GA and genetic operators work to optimize possible solutions. This study is later used as the baseline for the development of parallel techniques capable of exploring the identified points of parallelism that simply focus on accelerating the heavy duty fitness function to a full GPU accelerated GA. 

João Baúto works at Fundacao Champalimaud in Lisbon, Portugal. He implements high performance computing tools applied to neuroscience and cancer research.

Rui Ferreira Neves is a professor at Instituto Superior Técnico, Portugal. His research activity comprises evolutionary computation and pattern matching applied to the financial markets, sensor networks, embedded systems and mixed signal integrated circuits.

Nuno Horta is the Head of the Integrated Circuits Group, Instituto de Telecomunicacoes, Portugal. His reseach interests are mainly in analog and mixed-sgnal IC design, analog IC design automation, soft computing and data science.

Inhalt

Preface 7
Contents 9
Acronyms 12
1 Introduction 14
1.1 Motivation 15
1.2 Goals 15
1.3 Book Outline 16
References 16
2 Background 17
2.1 Time Series Analysis 17
2.1.1 Euclidean Distance 17
2.1.2 Dynamic Time Warping 18
2.1.3 Piecewise Linear Approximation 18
2.1.4 Piecewise Aggregate Approximation 19
2.1.5 Symbolic Aggregate approXimation 20
2.2 Genetic Algorithm 22
2.2.1 Selection Operator 23
2.2.2 Crossover Operator 24
2.2.3 Mutation Operator 24
2.3 Graphics Processing Units 25
2.3.1 NVIDIA's GPU Architecture Overview 25
2.3.2 NVIDIA's GPU Architectures 27
2.3.3 CUDA Architecture 29
2.4 Conclusions 31
References 31
3 State-of-the-Art in Pattern Recognition Techniques 33
3.1 Middle Curve Piecewise Linear Approximation 33
3.2 Perceptually Important Points 34
3.3 Turning Points 38
3.4 Symbolic Aggregate approXimation 40
3.5 Shapelets 40
3.6 Conclusions 42
References 43
4 SAX/GA CPU Approach 1
4.1 SAX/GA CPU Approach 45
4.1.1 Population Generation 46
4.1.2 Fitness Evaluation 46
4.1.3 Population Selection 50
4.1.4 Chromosome Crossover 50
4.1.5 Individual Mutation 51
4.2 SAX/GA Performance Analysis 52
4.3 Conclusions 55
References 56
5 GPU-Accelerated SAX/GA 57
5.1 Parallel SAX Representation 57
5.1.1 Prototype 1: SAX Transformation On-Demand 57
5.1.2 Prototype 2: Speculative FSM 59
5.1.3 Solution A: SAX/GA with Speculative GPU SAX Transformation 62
5.2 Parallel Dataset Training 67
5.2.1 Prototype 3: Parallel SAX/GA Training 67
5.2.2 Solution B: Parallel SAX/GA Training with GPU Fitness Evaluation 69
5.3 Fully GPU-Accelerated SAX/GA 72
5.3.1 Population Generation Kernel 73
5.3.2 Population Selection 74
5.3.3 Gene Crossover Kernel 75
5.3.4 Gene Mutation Kernel 75
5.3.5 Execution Flow 77
5.4 Conclusions 77
Reference 78
6 Results 79
6.1 SAX/GA Initial Constraints 79
6.2 Study Case A: Execution Time 80
6.2.1 Solution A: SAX/GA with Speculative FSM 80
6.2.2 Solution B: Parallel Dataset Training 86
6.2.3 Solution C: Fully GPU-Accelerated SAX/GA 89
6.3 Study Case B: FSM Prediction Rate 93
6.4 Study Case C: Quality of Solutions 96
6.5 Conclusions 100
7 Conclusions and Future Work 101
7.1 Future Work 103

Produkt-Details

• Erscheint lt. Verlag: 3.2.2018
• Reihe/Serie: SpringerBriefs in Applied Sciences and Technology
• Reihe/Serie: SpringerBriefs in Applied Sciences and Technology
• Reihe/Serie: SpringerBriefs in Computational Intelligence
• Reihe/Serie: SpringerBriefs in Computational Intelligence
• Zusatzinfo: XIV, 91 p. 50 illus.
• Verlagsort: Cham
• Sprache: englisch
• Themenwelt: Mathematik / Informatik ► Mathematik
• Themenwelt: Technik
• Schlagworte: Computational Finance • Data Science • High Performance Computing • market trading strategies • pattern recognition techniques • Quantitative Finance • SAX/GA algorithm
• ISBN-13: 9783319733296 / 9783319733296