Cargando…
Neurocomputing : learning, architectures, and modeling /
| Clasificación: | Libro Electrónico |
|---|---|
| Otros Autores: | |
| Formato: | Electrónico eBook |
| Idioma: | Inglés |
| Publicado: |
New York :
Nova Science Publishers, Inc.,
[2012]
|
| Colección: | Computing science, technology and applications.
|
| Temas: | |
| Acceso en línea: | Texto completo |
Tabla de Contenidos:
- NEUROCOMPUTING ; NEUROCOMPUTING ; CONTENTS ; PREFACE ; INTELLIGENT MARKET: A BRAIN-COMPUTER INTERFACE FOR ANALYZING INVESTMENT BEHAVIOR AND MARKET STABILITY ; ABSTRACT ; 1. INTRODUCTION ; 2. SYSTEM CON GURATION ; 2.1. Overview ; 2.2. Arti cial Market Component ; 2.2.1. Client Component ; 2.2.2. Server Component ; 2.3. Functional Brain Measurement and Real-Time Processing Component ; 2.3.1. fNIRS; 2.3.2. Speci cation of Measurement Sites ; 2.3.3. Real-Time Data Transfer and ltering ; 2.4. Construction of a Predictive Model of Investment Behavior and a Sequential Learning Component.
- 2.4.1. Predictive Factors 2.4.2. Ensemble Learning 1: Predicting Investment Behavior with an SVM; 2.4.3. Ensemble learning 2: Market Price Prediction by a Bayesian 3-Layer Percep-Tron ; (A) 3-Layer Perceptron ; (B) Prior Distribution of Parameters and Hyperparameters ; (C) Hyperparameter Marginal Likelihood ; (D) Mean Squared Deviation ; 3. ASSESSMENT OF THE SYSTEM ; 3.1. Experimental Procedures; 3.2. Investment Decisions; 3.3. Introduction of a CTA; 3.4. Performance Assessment ; 3.5. Market Price Prediction ; 4. CONCLUSION ; 1. APPENDIX A: BAYESIAN THREE-LAYER PERCEPTRON ; (i) Perceptron.
- (Ii) Prior Distributions (ii-a) Prior distribution of parameters k=(a, b, c) ; (ii-b) Prior distribution of hyperparameters Üc ; (ii-c) Prior distribution of hyperparameters . ; (iii) Likelihood ; (iv) Posteriori Distributions ; (iv-a) Conditional posterior distribution of parameters k=(a, b, c) ; (iv-b) Conditional posterior distribution of hyperparameter Üc ; (iv-c) Conditional posterior distribution of hyperparameters Ý ; (V) Markov Chain Monte Carlo Sampling Method ; (v-a) Proposal distributions for sampling bki and ck ; 2. APPENDIX B: MODEL SELECTION ; REFERENCES.
- NEURAL-BASED IMAGE SEGMENTATION ARCHITECTURE WITH EXECUTION ON A GPU ABSTRACT ; 1. INTRODUCTION ; 2. GPU STREAM PROCESSING MODEL ; 2.1. Mapping Neural Architectures to a Stream Processing Model ; 3. DESCRIPTION OF THE NEURAL ARCHITECTURE ; 4. COLOUR OPPONENCY (COP) ; 4.1. Type I and Type III Cells ; 4.2. Type II Cells ; 5. BOUNDARY DETECTION (BOD) ; 5.1. Simple Cells ; 5.2. Complex Cells ; 5.3. Competition and Cooperation ; 5.3.1. Competition ; 5.3.2. Cooperation ; 6. CHROMATIC DIFFUSION (CHD) ; 6.1. Chromatic Double Opponent Cells (CDOC) ; 6.2. Diffusion; 7. SCALE FUSION (SCF)
- 8. EXPERIMENTAL RESULTS 8.1. GPU Implementation Performance ; 8.2. Performance in Presence of Noise ; 8.3. Importance of Colour Opponency ; 8.4. Illusory Boundary Generation ; 8.5. Berkeley Segmentation Tests ; 9. CONCLUSION AND FUTURE WORK ; ACKNOWLEDGMENTS ; A. MODEL EQUATIONS ; A1. Colour Opponency (CoP) ; A1.1. Type I and Type III Cells ; A1.2. Type II cells ; A2. Boundary Detection (BoD) ; A2.1. Simple Cells ; A2.2. Complex Cells; A2.3. Competition ; A2.4. Cooperation ; A3. Chromatic Diffusion (ChD) ; A4. Scale fusion (ScF) ; REFERENCES.