Signal Processing and Machine Learning Theory, authored by world-leading experts, reviews the principles, methods and techniques of essential and advanced signal processing theory. These theories and tools are the driving engines of many current and emerging research topics and technologies, such as machine learning, autonomous vehicles, the internet of things, future wireless communications, medical imaging, etc.
1. Chapter 1 Introduction to signal processing and machine learning theory - p. 1 2. Chapter 2 Continuous-time signals and systems - p. 35 3. 2.1: Introduction - p. 35 4. 2.2: Continuous-time systems - p. 37 5. 2.3: Differential equations - p. 43 6. 2.4: Laplace transform: definition and properties - p. 48 7. 2.5: Transfer function and stability - p. 61 8. 2.6: Frequency response - p. 65 9. 2.7: The Fourier series and the Fourier transform - p. 72 10. 2.8: Conclusion and future trends - p. 82 11. Relevant websites - p. 82 12. Chapter 3 Discrete-time signals and systems - p. 85 13. 3.1: Introduction - p. 85 14. 3.2: Discrete-time signals: sequences - p. 85 15. 3.3: Discrete-time systems - p. 87 16. 3.4: Linear time-invariant systems - p. 89 17. 3.5: Discrete-time signals and systems with MATLAB® - p. 108 18. 3.6: Conclusions - p. 118 19. Notation - p. 118 20. Chapter 4 Random signals and stochastic processes - p. 123 21. Foreword - p. 123 22. Acknowledgment - p. 123 23. 4.1: Introduction - p. 123 24. 4.2: Probability - p. 124 25. 4.3: Random variable - p. 129 26. 4.3.1: Probability distributions - p. 131 27. 4.3.2: Usual distributions - p. 134 28. 4.3.3: Conditional distribution - p. 137 29. 4.3.4: Statistical moments - p. 137 30. 4.3.5: Transformation of random variables - p. 141 31. 4.3.6: Multiple random variable distributions - p. 142 32. 4.3.7: Statistically independent random variables - p. 146 33. 4.3.8: Joint statistical moments - p. 146 34. 4.3.9: Central limit theorem - p. 148 35. 4.3.10: Multivariate Gaussian distribution - p. 149 36. 4.3.11: Transformation of vector random variables - p. 149 37. 4.3.12: Complex random variables - p. 150 38. 4.3.13: Application: estimators - p. 151 39. 4.3.14: Application: feature selection - p. 153 40. 4.4: Random process - p. 153 41. 4.4.1: Distributions of random processes and sequences - p. 154 42. 4.4.2: Statistical independence - p. 155 43. 4.4.3: First- and second-order moments for random processes and sequences - p. 155 44. 4.4.4: Stationarity - p. 157 45. 4.4.5: Properties of correlation functions for WSS processes and sequences - p. 158 46. 4.4.6: Time averages of random processes and sequences - p. 159 47. 4.4.7: Ergodicity - p. 161 48. 4.4.8: An encompassing example - p. 162 49. 4.4.9: Gaussian processes and sequences - p. 165 50. 4.4.10: Poisson random process - p. 165 51. 4.4.11: Complex random processes and sequences - p. 166 52. 4.4.12: Markov chains - p. 167 53. 4.4.13: Spectral description of random processes and sequences - p. 169 54. 4.4.14: White and colored noise - p. 172 55. 4.4.15: Applications: modulation, “bandpass,” and band-limited processes, and sampling - p. 173 56. 4.4.16: Processing of random processes and sequences - p. 175 57. 4.4.17: Application: characterization of LTI systems - p. 179 58. 4.4.18: Modeling of bandpass WSS random processes - p. 180 59. 4.4.19: Statistical modeling of signals: random sequence as the output of an LTI system - p. 181 60. 4.5: Afterword - p. 183 61. Chapter 5 Sampling and quantization - p. 185 62. 5.1: Introduction - p. 185 63. 5.2: Preliminaries - p. 187 64. 5.3: Sampling of deterministic signals - p. 194 65. 5.4: Sampling of stochastic processes - p. 212 66. 5.5: Nonuniform sampling and generalizations - p. 220 67. 5.6: Quantization - p. 234 68. 5.7: Oversampling techniques and MIMO systems - p. 244 69. 5.8: Discrete-time modeling of mixed-signal systems - p. 251 70. Chapter 6 Digital filter structures and their implementation - p. 267 71. 6.1: Introduction - p. 267 72. 6.2: Properties of digital filters - p. 270 73. 6.3: Synthesis of digital FIR filters - p. 272 74. 6.4: FIR structures - p. 279 75. 6.5: Frequency response masking filters - p. 285 76. 6.6: The analog approximation problem - p. 316 77. 6.7: Doubly resistively terminated lossless networks - p. 321 78. 6.8: Design of doubly resistively terminated analog filters - p. 327 79. 6.9: Design and realization of IIR filters - p. 338 80. 6.10: Wave digital filters - p. 339 81. 6.11: Ladder wave digital filters - p. 359 82. 6.12: Design of lattice wave digital filters - p. 361 83. 6.13: Circulator-tree wave digital filters - p. 373 84. 6.14: Numerically equivalent state-space realization of wave digital filters - p. 377 85. 6.15: Computational properties of filter algorithms - p. 378 86. 6.16: Architecture - p. 380 87. 6.17: Arithmetic operations - p. 381 88. 6.18: Composite arithmetic operations - p. 388 89. 6.19: Power reduction techniques - p. 393 90. Chapter 7 Multirate signal processing for software radio architectures - p. 403 91. 7.1: Introduction - p. 403 92. 7.2: The sampling process and the “resampling” process - p. 407 93. 7.3: Digital filters - p. 413 94. 7.4: Windowing - p. 423 95. 7.5: Basics on multirate filters - p. 426 96. 7.6: From single-channel down converter to standard down converter channelizer - p. 427 97. 7.7: Modifications of the standard down converter channelizer – M:2 down converter channelizer - p. 444 98. 7.8: Preliminaries on software-defined radios - p. 460 99. 7.9: Proposed architectures for software radios - p. 462 100. 7.10: Case study: enabling automated signal detection, segregation, and classification - p. 479 101. 7.11: Closing comments - p. 492 102. Glossary - p. 493 103. Chapter 8 Modern transform design for practical audio/image/video coding applications - p. 495 104. 8.1: Introduction - p. 495 105. 8.2: Background and fundamentals - p. 497 106. 8.3: Design strategy - p. 503 107. 8.4: Approximation approach via direct scaling - p. 505 108. 8.5: Approximation approach via structural design - p. 508 109. 8.6: Wavelet filter design via spectral factorization - p. 516 110. 8.7: Higher-order design approach via optimization - p. 522 111. 8.8: Conclusion - p. 534 112. Chapter 9 Data representation: from multiscale transforms to neural networks - p. 537 113. 9.1: Introduction - p. 537 114. 9.2: Wavelets: a multiscale analysis tool - p. 545 115. 9.3: Curvelets and their applications - p. 567 116. 9.4: Contourlets and their applications - p. 586 117. 9.5: Shearlets and their applications - p. 605 118. 9.6: Incorporating wavelets into neural networks - p. 625 119. Appendix 9.A: - p. 643 120. Chapter 10 Frames in signal processing - p. 651 121. 10.1: Introduction - p. 651 122. 10.2: Basic concepts - p. 652 123. 10.3: Relevant definitions - p. 656 124. 10.4: Some computational remarks - p. 664 125. 10.5: Construction of frames from a prototype signal - p. 667 126. 10.6: Some remarks and highlights on applications - p. 673 127. 10.7: Conclusion - p. 684 128. Chapter 11 Parametric estimation - p. 689 129. 11.1: Introduction - p. 689 130. 11.2: Preliminaries - p. 692 131. 11.3: Parametric models for linear time-invariant systems - p. 696 132. 11.4: Joint process estimation and sequential modeling - p. 707 133. 11.5: Model order estimation - p. 714 134. Chapter 12 Adaptive filters - p. 717 135. 12.1: Introduction - p. 717 136. 12.1.1: Motivation – acoustic echo cancelation - p. 718 137. 12.1.2: A quick tour of adaptive filtering - p. 721 138. 12.1.3: Applications - p. 740 139. 12.2: Optimum filtering - p. 745 140. 12.2.1: Linear least-mean squares estimation - p. 746 141. 12.2.2: Complex variables and multichannel filtering - p. 760 142. 12.3: Stochastic algorithms - p. 773 143. 12.4: Statistical analysis - p. 790 144. 12.4.1: Data model - p. 792 145. 12.4.2: Relating the autocorrelation matrix of the weight error vector to the EMSE and the MSD - p. 793 146. 12.4.3: Statistical analysis of the LMS algorithm - p. 796 147. 12.4.4: A unified statistical analysis - p. 801 148. 12.5: Extensions and current research - p. 818 149. 12.5.1: Finite precision arithmetic - p. 819 150. 12.5.2: Regularization - p. 830 151. 12.5.3: Variable step size - p. 832 152. 12.5.4: Non-Gaussian noise and robust filters - p. 833 153. 12.5.5: Blind equalization - p. 835 154. 12.5.6: Subband and transform-domain adaptive filters - p. 837 155. 12.5.7: Affine projections algorithm - p. 837 156. 12.5.8: Cooperative estimation - p. 839 157. 12.5.9: Adaptive IIR filters - p. 845 158. 12.5.10: Set membership and projections onto convex sets - p. 846 159. 12.5.11: Adaptive filters with constraints - p. 847 160. 12.5.12: Reduced-rank adaptive filters - p. 848 161. 12.5.13: Kernel adaptive filtering - p. 848 162. Chapter 13 Machine learning - p. 869 163. 13.1: Introduction - p. 869 164. 13.2: Learning concepts - p. 870 165. 13.3: Unsupervised learning - p. 876 166. 13.4: Supervised learning - p. 879 167. 13.4.1: Perceptron - p. 879 168. 13.4.2: Fully connected neural networks - p. 883 169. 13.4.3: Regression - p. 887 170. 13.4.4: A regression example - p. 889 171. 13.4.5: Classification - p. 892 172. 13.4.6: A classification example - p. 892 173. 13.4.7: EMNIST letter dataset - p. 893 174. 13.4.8: Dropout - p. 895 175. 13.4.9: Number of flops - p. 897 176. 13.4.10: Convolutional neural networks - p. 898 177. 13.4.11: Recurrent neural networks - p. 899 178. 13.4.12: Support vector machines - p. 908 179. 13.5: Ensemble learning - p. 921 180. 13.6: Deep learning - p. 924 181. 13.7: CNN visualization - p. 936 182. 13.8: Deep reinforcement learning - p. 938 183. 13.9: Current trends - p. 941 184. 13.10: Concluding remarks - p. 949 185. 13.11: Appendix: RNN's gradient derivations - p. 949 186. Chapter 14 A primer on graph signal processing - p. 961 187. 14.1: The case for GSP - p. 961 188. 14.2: Fundamentals of graph theory - p. 962 189. 14.3: Graph signals and systems - p. 965 190. 14.4: Graph Fourier transform - p. 969 191. 14.5: Graph filtering - p. 976 192. 14.6: Down- and upsampling graph signals - p. 984 193. 14.7: Examples and applications - p. 993 194. 14.8: A short history of GSP - p. 1001 195. Chapter 15 Tensor methods in deep learning - p. 1009 196. 15.1: Introduction - p. 1009 197. 15.2: Preliminaries on matrices and tensors - p. 1011 198. 15.3: Tensor decomposition - p. 1014 199. 15.4: Tensor methods in deep learning architectures - p. 1019 200. 15.5: Tensor methods in quantum machine learning - p. 1034 201. 15.6: Software - p. 1037 202. 15.7: Limitations and guidelines - p. 1038 203. Chapter 16 Nonconvex graph learning: sparsity, heavy tails, and clustering - p. 1049 204. Chapter 17 Dictionaries in machine learning - p. 1073 205. 17.1: Data-driven AI via dictionary learning for sparse signal processing - p. 1073 206. 17.2: Sparsity and sparse representations - p. 1075 207. 17.3: Sparse signal processing - p. 1082 208. 17.4: Dictionary learning I – basic models and algorithms - p. 1086 209. 17.5: Dictionary learning II – the hierarchical/empirical Bayes approach - p. 1101 210. 17.6: Nonnegative matrix factorization in dictionary learning - p. 1116 211. 17.7: Dictionaries, data manifold learning, and geometric multiresolution analysis - p. 1119 212. 17.8: Hard clustering and classification in dictionary learning - p. 1122 213. 17.9: Multilayer dictionary learning and classification - p. 1127 214. 17.10: Kernel dictionary learning - p. 1131 215. 17.11: Conclusion - p. 1136 216. Appendix 17.A: Derivation and properties of the K-SVD algorithm - p. 1137 217. Appendix 17.B: Derivation of the SBL EM update equation - p. 1141 218. Appendix 17.C: SBL dictionary learning algorithm - p. 1142 219. Appendix 17.D: Mathematical background for kernelizing dictionary learning - p. 1146 220. Index - p. 1161
Signal Processing and Machine Learning Theory, authored by world-leading experts, reviews the principles, methods and techniques of essential and advanced signal processing theory. These theories and tools are the driving engines of many current and emerging research topics and technologies, such as machine learning, autonomous vehicles, the internet of things, future wireless communications, medical imaging, etc.
1. Chapter 1 Introduction to signal processing and machine learning theory - p. 1 2. Chapter 2 Continuous-time signals and systems - p. 35 3. 2.1: Introduction - p. 35 4. 2.2: Continuous-time systems - p. 37 5. 2.3: Differential equations - p. 43 6. 2.4: Laplace transform: definition and properties - p. 48 7. 2.5: Transfer function and stability - p. 61 8. 2.6: Frequency response - p. 65 9. 2.7: The Fourier series and the Fourier transform - p. 72 10. 2.8: Conclusion and future trends - p. 82 11. Relevant websites - p. 82 12. Chapter 3 Discrete-time signals and systems - p. 85 13. 3.1: Introduction - p. 85 14. 3.2: Discrete-time signals: sequences - p. 85 15. 3.3: Discrete-time systems - p. 87 16. 3.4: Linear time-invariant systems - p. 89 17. 3.5: Discrete-time signals and systems with MATLAB® - p. 108 18. 3.6: Conclusions - p. 118 19. Notation - p. 118 20. Chapter 4 Random signals and stochastic processes - p. 123 21. Foreword - p. 123 22. Acknowledgment - p. 123 23. 4.1: Introduction - p. 123 24. 4.2: Probability - p. 124 25. 4.3: Random variable - p. 129 26. 4.3.1: Probability distributions - p. 131 27. 4.3.2: Usual distributions - p. 134 28. 4.3.3: Conditional distribution - p. 137 29. 4.3.4: Statistical moments - p. 137 30. 4.3.5: Transformation of random variables - p. 141 31. 4.3.6: Multiple random variable distributions - p. 142 32. 4.3.7: Statistically independent random variables - p. 146 33. 4.3.8: Joint statistical moments - p. 146 34. 4.3.9: Central limit theorem - p. 148 35. 4.3.10: Multivariate Gaussian distribution - p. 149 36. 4.3.11: Transformation of vector random variables - p. 149 37. 4.3.12: Complex random variables - p. 150 38. 4.3.13: Application: estimators - p. 151 39. 4.3.14: Application: feature selection - p. 153 40. 4.4: Random process - p. 153 41. 4.4.1: Distributions of random processes and sequences - p. 154 42. 4.4.2: Statistical independence - p. 155 43. 4.4.3: First- and second-order moments for random processes and sequences - p. 155 44. 4.4.4: Stationarity - p. 157 45. 4.4.5: Properties of correlation functions for WSS processes and sequences - p. 158 46. 4.4.6: Time averages of random processes and sequences - p. 159 47. 4.4.7: Ergodicity - p. 161 48. 4.4.8: An encompassing example - p. 162 49. 4.4.9: Gaussian processes and sequences - p. 165 50. 4.4.10: Poisson random process - p. 165 51. 4.4.11: Complex random processes and sequences - p. 166 52. 4.4.12: Markov chains - p. 167 53. 4.4.13: Spectral description of random processes and sequences - p. 169 54. 4.4.14: White and colored noise - p. 172 55. 4.4.15: Applications: modulation, “bandpass,” and band-limited processes, and sampling - p. 173 56. 4.4.16: Processing of random processes and sequences - p. 175 57. 4.4.17: Application: characterization of LTI systems - p. 179 58. 4.4.18: Modeling of bandpass WSS random processes - p. 180 59. 4.4.19: Statistical modeling of signals: random sequence as the output of an LTI system - p. 181 60. 4.5: Afterword - p. 183 61. Chapter 5 Sampling and quantization - p. 185 62. 5.1: Introduction - p. 185 63. 5.2: Preliminaries - p. 187 64. 5.3: Sampling of deterministic signals - p. 194 65. 5.4: Sampling of stochastic processes - p. 212 66. 5.5: Nonuniform sampling and generalizations - p. 220 67. 5.6: Quantization - p. 234 68. 5.7: Oversampling techniques and MIMO systems - p. 244 69. 5.8: Discrete-time modeling of mixed-signal systems - p. 251 70. Chapter 6 Digital filter structures and their implementation - p. 267 71. 6.1: Introduction - p. 267 72. 6.2: Properties of digital filters - p. 270 73. 6.3: Synthesis of digital FIR filters - p. 272 74. 6.4: FIR structures - p. 279 75. 6.5: Frequency response masking filters - p. 285 76. 6.6: The analog approximation problem - p. 316 77. 6.7: Doubly resistively terminated lossless networks - p. 321 78. 6.8: Design of doubly resistively terminated analog filters - p. 327 79. 6.9: Design and realization of IIR filters - p. 338 80. 6.10: Wave digital filters - p. 339 81. 6.11: Ladder wave digital filters - p. 359 82. 6.12: Design of lattice wave digital filters - p. 361 83. 6.13: Circulator-tree wave digital filters - p. 373 84. 6.14: Numerically equivalent state-space realization of wave digital filters - p. 377 85. 6.15: Computational properties of filter algorithms - p. 378 86. 6.16: Architecture - p. 380 87. 6.17: Arithmetic operations - p. 381 88. 6.18: Composite arithmetic operations - p. 388 89. 6.19: Power reduction techniques - p. 393 90. Chapter 7 Multirate signal processing for software radio architectures - p. 403 91. 7.1: Introduction - p. 403 92. 7.2: The sampling process and the “resampling” process - p. 407 93. 7.3: Digital filters - p. 413 94. 7.4: Windowing - p. 423 95. 7.5: Basics on multirate filters - p. 426 96. 7.6: From single-channel down converter to standard down converter channelizer - p. 427 97. 7.7: Modifications of the standard down converter channelizer – M:2 down converter channelizer - p. 444 98. 7.8: Preliminaries on software-defined radios - p. 460 99. 7.9: Proposed architectures for software radios - p. 462 100. 7.10: Case study: enabling automated signal detection, segregation, and classification - p. 479 101. 7.11: Closing comments - p. 492 102. Glossary - p. 493 103. Chapter 8 Modern transform design for practical audio/image/video coding applications - p. 495 104. 8.1: Introduction - p. 495 105. 8.2: Background and fundamentals - p. 497 106. 8.3: Design strategy - p. 503 107. 8.4: Approximation approach via direct scaling - p. 505 108. 8.5: Approximation approach via structural design - p. 508 109. 8.6: Wavelet filter design via spectral factorization - p. 516 110. 8.7: Higher-order design approach via optimization - p. 522 111. 8.8: Conclusion - p. 534 112. Chapter 9 Data representation: from multiscale transforms to neural networks - p. 537 113. 9.1: Introduction - p. 537 114. 9.2: Wavelets: a multiscale analysis tool - p. 545 115. 9.3: Curvelets and their applications - p. 567 116. 9.4: Contourlets and their applications - p. 586 117. 9.5: Shearlets and their applications - p. 605 118. 9.6: Incorporating wavelets into neural networks - p. 625 119. Appendix 9.A: - p. 643 120. Chapter 10 Frames in signal processing - p. 651 121. 10.1: Introduction - p. 651 122. 10.2: Basic concepts - p. 652 123. 10.3: Relevant definitions - p. 656 124. 10.4: Some computational remarks - p. 664 125. 10.5: Construction of frames from a prototype signal - p. 667 126. 10.6: Some remarks and highlights on applications - p. 673 127. 10.7: Conclusion - p. 684 128. Chapter 11 Parametric estimation - p. 689 129. 11.1: Introduction - p. 689 130. 11.2: Preliminaries - p. 692 131. 11.3: Parametric models for linear time-invariant systems - p. 696 132. 11.4: Joint process estimation and sequential modeling - p. 707 133. 11.5: Model order estimation - p. 714 134. Chapter 12 Adaptive filters - p. 717 135. 12.1: Introduction - p. 717 136. 12.1.1: Motivation – acoustic echo cancelation - p. 718 137. 12.1.2: A quick tour of adaptive filtering - p. 721 138. 12.1.3: Applications - p. 740 139. 12.2: Optimum filtering - p. 745 140. 12.2.1: Linear least-mean squares estimation - p. 746 141. 12.2.2: Complex variables and multichannel filtering - p. 760 142. 12.3: Stochastic algorithms - p. 773 143. 12.4: Statistical analysis - p. 790 144. 12.4.1: Data model - p. 792 145. 12.4.2: Relating the autocorrelation matrix of the weight error vector to the EMSE and the MSD - p. 793 146. 12.4.3: Statistical analysis of the LMS algorithm - p. 796 147. 12.4.4: A unified statistical analysis - p. 801 148. 12.5: Extensions and current research - p. 818 149. 12.5.1: Finite precision arithmetic - p. 819 150. 12.5.2: Regularization - p. 830 151. 12.5.3: Variable step size - p. 832 152. 12.5.4: Non-Gaussian noise and robust filters - p. 833 153. 12.5.5: Blind equalization - p. 835 154. 12.5.6: Subband and transform-domain adaptive filters - p. 837 155. 12.5.7: Affine projections algorithm - p. 837 156. 12.5.8: Cooperative estimation - p. 839 157. 12.5.9: Adaptive IIR filters - p. 845 158. 12.5.10: Set membership and projections onto convex sets - p. 846 159. 12.5.11: Adaptive filters with constraints - p. 847 160. 12.5.12: Reduced-rank adaptive filters - p. 848 161. 12.5.13: Kernel adaptive filtering - p. 848 162. Chapter 13 Machine learning - p. 869 163. 13.1: Introduction - p. 869 164. 13.2: Learning concepts - p. 870 165. 13.3: Unsupervised learning - p. 876 166. 13.4: Supervised learning - p. 879 167. 13.4.1: Perceptron - p. 879 168. 13.4.2: Fully connected neural networks - p. 883 169. 13.4.3: Regression - p. 887 170. 13.4.4: A regression example - p. 889 171. 13.4.5: Classification - p. 892 172. 13.4.6: A classification example - p. 892 173. 13.4.7: EMNIST letter dataset - p. 893 174. 13.4.8: Dropout - p. 895 175. 13.4.9: Number of flops - p. 897 176. 13.4.10: Convolutional neural networks - p. 898 177. 13.4.11: Recurrent neural networks - p. 899 178. 13.4.12: Support vector machines - p. 908 179. 13.5: Ensemble learning - p. 921 180. 13.6: Deep learning - p. 924 181. 13.7: CNN visualization - p. 936 182. 13.8: Deep reinforcement learning - p. 938 183. 13.9: Current trends - p. 941 184. 13.10: Concluding remarks - p. 949 185. 13.11: Appendix: RNN's gradient derivations - p. 949 186. Chapter 14 A primer on graph signal processing - p. 961 187. 14.1: The case for GSP - p. 961 188. 14.2: Fundamentals of graph theory - p. 962 189. 14.3: Graph signals and systems - p. 965 190. 14.4: Graph Fourier transform - p. 969 191. 14.5: Graph filtering - p. 976 192. 14.6: Down- and upsampling graph signals - p. 984 193. 14.7: Examples and applications - p. 993 194. 14.8: A short history of GSP - p. 1001 195. Chapter 15 Tensor methods in deep learning - p. 1009 196. 15.1: Introduction - p. 1009 197. 15.2: Preliminaries on matrices and tensors - p. 1011 198. 15.3: Tensor decomposition - p. 1014 199. 15.4: Tensor methods in deep learning architectures - p. 1019 200. 15.5: Tensor methods in quantum machine learning - p. 1034 201. 15.6: Software - p. 1037 202. 15.7: Limitations and guidelines - p. 1038 203. Chapter 16 Nonconvex graph learning: sparsity, heavy tails, and clustering - p. 1049 204. Chapter 17 Dictionaries in machine learning - p. 1073 205. 17.1: Data-driven AI via dictionary learning for sparse signal processing - p. 1073 206. 17.2: Sparsity and sparse representations - p. 1075 207. 17.3: Sparse signal processing - p. 1082 208. 17.4: Dictionary learning I – basic models and algorithms - p. 1086 209. 17.5: Dictionary learning II – the hierarchical/empirical Bayes approach - p. 1101 210. 17.6: Nonnegative matrix factorization in dictionary learning - p. 1116 211. 17.7: Dictionaries, data manifold learning, and geometric multiresolution analysis - p. 1119 212. 17.8: Hard clustering and classification in dictionary learning - p. 1122 213. 17.9: Multilayer dictionary learning and classification - p. 1127 214. 17.10: Kernel dictionary learning - p. 1131 215. 17.11: Conclusion - p. 1136 216. Appendix 17.A: Derivation and properties of the K-SVD algorithm - p. 1137 217. Appendix 17.B: Derivation of the SBL EM update equation - p. 1141 218. Appendix 17.C: SBL dictionary learning algorithm - p. 1142 219. Appendix 17.D: Mathematical background for kernelizing dictionary learning - p. 1146 220. Index - p. 1161
