Muutke küpsiste eelistusi

Deep Learning through Sparse and Low-Rank Modeling [Pehme köide]

3.00/5 (2 hinnangut Goodreads-ist)
(Professor, University of Illinois at Urbana-Champaign, USA), (Associate Professor, Northeastern University, USA), (Assistant Professor, Texas A&M University, USA)
  • Formaat: Paperback / softback, 296 pages, kõrgus x laius: 235x191 mm, kaal: 570 g
  • Sari: Computer Vision and Pattern Recognition
  • Ilmumisaeg: 12-Apr-2019
  • Kirjastus: Academic Press Inc
  • ISBN-10: 0128136596
  • ISBN-13: 9780128136591
Teised raamatud teemal:
Deep Learning through Sparse and Low-Rank ModelingSuurem pilt
  • Formaat: Paperback / softback, 296 pages, kõrgus x laius: 235x191 mm, kaal: 570 g
  • Sari: Computer Vision and Pattern Recognition
  • Ilmumisaeg: 12-Apr-2019
  • Kirjastus: Academic Press Inc
  • ISBN-10: 0128136596
  • ISBN-13: 9780128136591
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9780128136591.html
Märksõnad:

Deep Learning Through Sparse Representation and Low-Rank Modeling bridges classical sparse and low rank models-those that emphasize problem-specific Interpretability-with recent deep network models that have enabled a larger learning capacity and better utilization of Big Data. It shows how the toolkit of deep learning is closely tied with the sparse/low rank methods and algorithms, providing a rich variety of theoretical and analytic tools to guide the design and interpretation of deep learning models. The development of the theory and models is supported by a wide variety of applications in computer vision, machine learning, signal processing, and data mining.

This book will be highly useful for researchers, graduate students and practitioners working in the fields of computer vision, machine learning, signal processing, optimization and statistics.

  • Combines classical sparse and low-rank models and algorithms with the latest advances in deep learning networks
  • Shows how the structure and algorithms of sparse and low-rank methods improves the performance and interpretability of Deep Learning models
  • Provides tactics on how to build and apply customized deep learning models for various applications
Contributors xi
About the Editors xiii
Preface xv
Acknowledgments xvii
1 Introduction
1(8)
Zhangyang Wang
Ding Liu
1.1 Basics of Deep Learning
1(1)
1.2 Basics of Sparsity and Low-Rankness
2(1)
1.3 Connecting Deep Learning to Sparsity and Low-Rankness
3(1)
1.4 Organization
4(5)
References
4(5)
2 Bi-Level Sparse Coding: A Hyperspectral Image Classification Example
9(22)
Zhangyang Wang
2.1 Introduction
9(3)
2.2 Formulation and Algorithm
12(5)
2.2.1 Notations
12(1)
2.2.2 Joint Feature Extraction and Classification
12(3)
2.2.3 Bi-level Optimization Formulation
15(1)
2.2.4 Algorithm
15(2)
2.3 Experiments
17(9)
2.3.1 Classification Performance on AVIRIS Indiana Pines Data
20(3)
2.3.2 Classification Performance on AVIRIS Salinas Data
23(1)
2.3.3 Classification Performance on University of Pavia Data
23(3)
2.4 Conclusion
26(1)
2.5 Appendix
27(4)
References
28(3)
3 Deep l0 Encoders: A Model Unfolding Example
31(16)
Zhangyang Wang
3.1 Introduction
31(1)
3.2 Related Work
32(2)
3.2.1 l0-and l1-Based Sparse Approximations
32(1)
3.2.2 Network Implementation of l1-Approximation
33(1)
3.3 Deep l0 Encoders
34(3)
3.3.1 Deep lo-Regularized Encoder
34(2)
3.3.2 Deep M-Sparse l0 Encoder
36(1)
3.3.3 Theoretical Properties
37(1)
3.4 Task-Driven Optimization
37(1)
3.5 Experiment
38(5)
3.5.1 Implementation
38(1)
3.5.2 Simulation on l0 Sparse Approximation
38(2)
3.5.3 Applications on Classification
40(2)
3.5.4 Applications on Clustering
42(1)
3.6 Conclusions and Discussions on Theoretical Properties
43(4)
References
44(3)
4 Single Image Super-Resolution: From Sparse Coding to Deep Learning
47(40)
Ding Liu
Thomas S. Huang
4.1 Robust Single Image Super-Resolution via Deep Networks with Sparse Prior
47(26)
4.1.1 Introduction
47(2)
4.1.2 Related Work
49(1)
4.1.3 Sparse Coding Based Network for Image SR
50(4)
4.1.4 Network Cascade for Scalable SR
54(3)
4.1.5 Robust SR for Real Scenarios
57(3)
4.1.6 Implementation Details
60(1)
4.1.7 Experiments
61(8)
4.1.8 Subjective Evaluation
69(3)
4.1.9 Conclusion and Future Work
72(1)
4.2 Learning a Mixture of Deep Networks for Single Image Super-Resolution
73(14)
4.2.1 Introduction
73(1)
4.2.2 The Proposed Method
74(2)
4.2.3 Implementation Details
76(1)
4.2.4 Experimental Results
77(4)
4.2.5 Conclusion and Future Work
81(2)
References
83(4)
5 From Bi-Level Sparse Clustering to Deep Clustering
87(34)
Zhangyang Wang
5.1 A Joint Optimization Framework of Sparse Coding and Discriminative Clustering
87(14)
5.1.1 Introduction
87(1)
5.1.2 Model Formulation
88(2)
5.1.3 Clustering-Oriented Cost Functions
90(3)
5.1.4 Experiments
93(5)
5.1.5 Conclusion
98(1)
5.1.6 Appendix
99(2)
5.2 Learning a Task-Specific Deep Architecture for Clustering
101(20)
5.2.1 Introduction
101(1)
5.2.2 Related Work
102(1)
5.2.3 Model Formulation
103(3)
5.2.4 A Deeper Look: Hierarchical Clustering by DTAGnet
106(1)
5.2.5 Experiment Results
107(10)
5.2.6 Conclusion
117(1)
References
117(4)
6 Signal Processing
121(22)
Zhangyang Wang
Ding Liu
Thomas S. Huang
6.1 Deeply Optimized Compressive Sensing
121(9)
6.1.1 Background
121(1)
6.1.2 An End-to-End Optimization Model of CS
122(2)
6.1.3 DOCS: Feed-Forward and Jointly Optimized CS
124(2)
6.1.4 Experiments
126(3)
6.1.5 Conclusion
129(1)
6.2 Deep Learning for Speech Denoising
130(13)
6.2.1 Introduction
130(1)
6.2.2 Neural Networks for Spectral Denoising
131(3)
6.2.3 Experimental Results
134(5)
6.2.4 Conclusion and Future Work
139(1)
References
140(3)
7 Dimensionality Reduction
143(40)
Shuyang Wang
Zhangyang Wang
Yun Fu
7.1 Marginalized Denoising Dictionary Learning with Locality Constraint
143(22)
7.1.1 Introduction
143(2)
7.1.2 Related Works
145(2)
7.1.3 Marginalized Denoising Dictionary Learning with Locality Constraint
147(10)
7.1.4 Experiments
157(7)
7.1.5 Conclusion
164(1)
7.1.6 Future Works
165(1)
7.2 Learning a Deep l∞ Encoder for Hashing
165(18)
7.2.1 Introduction
166(2)
7.2.2 ADMM Algorithm
168(1)
7.2.3 Deep l∞ Encoder
168(2)
7.2.4 Deep l∞ Siamese Network for Hashing
170(2)
7.2.5 Experiments in Image Hashing
172(6)
7.2.6 Conclusion
178(1)
References
178(5)
8 Action Recognition
183(30)
Yu Kong
Yun Fu
8.1 Deeply Learned View-Invariant Features for Cross-View Action Recognition
183(15)
8.1.1 Introduction
183(2)
8.1.2 Related Work
185(1)
8.1.3 Deeply Learned View-Invariant Features
186(5)
8.1.4 Experiments
191(7)
8.2 Hybrid Neural Network for Action Recognition from Depth Cameras
198(11)
8.2.1 Introduction
198(1)
8.2.2 Related Work
199(2)
8.2.3 Hybrid Convolutional-Recursive Neural Networks
201(5)
8.2.4 Experiments
206(3)
8.3 Summary
209(4)
References
210(3)
9 Style Recognition and Kinship Understanding
213(38)
Shuhui Jiang
Ming Shao
Caiming Xiong
Yun Fu
9.1 Style Classification by Deep Learning
213(17)
9.1.1 Background
213(4)
9.1.2 Preliminary Knowledge of Stacked Autoencoder (SAE)
217(1)
9.1.3 Style Centralizing Autoencoder
217(4)
9.1.4 Consensus Style Centralizing Autoencoder
221(5)
9.1.5 Experiments
226(4)
9.2 Visual Kinship Understanding
230(16)
9.2.1 Background
230(2)
9.2.2 Related Work
232(1)
9.2.3 Family Faces
233(1)
9.2.4 Regularized Parallel Autoencoders
234(5)
9.2.5 Experimental Results
239(7)
9.3 Research Challenges and Future Works
246(5)
References
246(5)
10 Image Dehazing: Improved Techniques
251(12)
Yu Liu
Guanlong Zhao
Boyuan Gong
Yang Li
Ritu Raj
Niraj Goel
Satya Kesav
Sandeep Gottimukkala
Zhangyang Wang
Wenqi Ren
Dacheng Tao
10.1 Introduction
251(1)
10.2 Review and Task Description
252(2)
10.2.1 Haze Modeling and Dehazing Approaches
253(1)
10.2.2 RESIDE Dataset
253(1)
10.3 Task 1: Dehazing as Restoration
254(3)
10.4 Task 2: Dehazing for Detection
257(3)
10.4.1 Solution Set 1: Enhancing Dehazing and/or Detection Modules in the Cascade
257(1)
10.4.2 Solution Set 2: Domain-Adaptive Mask-RCNN
257(3)
10.5 Conclusion
260(3)
References
261(2)
11 Biomedical Image Analytics: Automated Lung Cancer Diagnosis
263(10)
Steve Kommrusch
Louis-Noel Pouchet
11.1 Introduction
263(1)
11.2 Related Work
264(1)
11.3 Methodology
265(3)
11.4 Experiments
268(2)
11.5 Conclusion
270(3)
Acknowledgments
271(1)
References
271(2)
Index 273
Dr. Zhangyang (Atlas) Wang is an Assistant Professor of Computer Science and Engineering (CSE), at the Texas A&M University (TAMU), since August 2017. During 2012-2016, he was a Ph.D. student in the Electrical and Computer Engineering (ECE) Department, at the University of Illinois at Urbana-Champaign (UIUC). He was a former research intern with Microsoft Research (2015), Adobe Research (2014), and US Army Research Lab (2013). Dr. Wang has published over 70 papers in top-tier venues, in the broad fields of machine learning, computer vision, artificial intelligence, and interdisciplinary data science. He has published 2 books and 1 chapter, has been granted 3 patents, and has received over 20 research awards and scholarships. Dr. Wang regularly serves as tutorial speakers, guest editors, area chairs, session chairs, TPC members, and workshop organizers at leading conferences and journals. Dr. Fu is an interdisciplinary faculty member affiliated with College of Engineering and the College of Computer and Information Science at Northeastern University. He received the B.Eng. degree in information engineering and the M.Eng. degree in pattern recognition and intelligence systems from Xi'an Jiaotong University, China, respectively, and the M.S. degree in statistics and the Ph.D. degree in electrical and computer engineering from the University of Illinois at Urbana-Champaign, respectively. Dr. Fu's research interests are Interdisciplinary research in Machine Learning and Computational Intelligence, Social Media Analytics, Human-Computer Interaction, and Cyber-Physical Systems. He has extensive publications in leading journals, books/book chapters and international conferences/workshops. Thomas S. Huang received his B.S. Degree in Electrical Engineering from National Taiwan University, Taipei, Taiwan, China; and his M.S. and Sc.D. Degrees in Electrical Engineering from the Massachusetts Institute of Technology, Cambridge, Massachusetts. He was on the Faculty of the Department of Electrical Engineering at MIT from 1963 to 1973; and on the Faculty of the School of Electrical Engineering and Director of its Laboratory for Information and Signal Processing at Purdue University from 1973 to 1980. Dr. Huang's professional interests lie in the broad area of information technology, especially the transmission and processing of multidimensional signals. He has published 21 books, and over 600 papers in Network Theory, Digital Filtering, Image Processing, and Computer Vision. Among his many honors and awards: Honda Lifetime Achievement Award, IEEE Jack Kilby Signal Processing Medal, and the King-Sun Fu Prize of the International Association for Pattern Recognition.