Advances in Deep Learning. (2019)

• Record Type: Book
• Further Information: Note: M. Arif Wani, Farooq Ahmad Bhat, Saduf Afzal, Asif Iqbal Khan.
• Other Names: Wani, M. A (M. Arif); Bhat, Farooq Ahmad; Afzal, Saduf; Khan, Asif Iqbal
• Contents: Intro; Preface; Contents; About the Authors; Abbreviations; 1 Introduction to Deep Learning; 1.1 Introduction; 1.2 Shallow Learning; 1.3 Deep Learning; 1.4 Why to Use Deep Learning; 1.5 How Deep Learning Works; 1.6 Deep Learning Challenges; Bibliography; 2 Basics of Supervised Deep Learning; 2.1 Introduction; 2.2 Convolutional Neural Network (ConvNet/CNN); 2.3 Evolution of Convolutional Neural Network Models; 2.4 Convolution Operation; 2.5 Architecture of CNN; 2.5.1 Convolution Layer; 2.5.2 Activation Function (ReLU); 2.5.3 Pooling Layer; 2.5.4 Fully Connected Layer; 2.5.5 Dropout 2.6 Challenges and Future Research DirectionBibliography; 3 Training Supervised Deep Learning Networks; 3.1 Introduction; 3.2 Training Convolution Neural Networks; 3.3 Loss Functions and Softmax Classifier; 3.3.1 Mean Squared Error (L2) Loss; 3.3.2 Cross-Entropy Loss; 3.3.3 Softmax Classifier; 3.4 Gradient Descent-Based Optimization Techniques; 3.4.1 Gradient Descent Variants; 3.4.2 Improving Gradient Descent for Faster Convergence; 3.5 Challenges in Training Deep Networks; 3.5.1 Vanishing Gradient; 3.5.2 Training Data Size; 3.5.3 Overfitting and Underfitting; 3.5.4 High-Performance Hardware 3.6 Weight Initialization Techniques3.6.1 Initialize All Weights to 0; 3.6.2 Random Initialization; 3.6.3 Random Weights from Probability Distribution; 3.6.4 Transfer Learning; 3.7 Challenges and Future Research Direction; Bibliography; 4 Supervised Deep Learning Architectures; 4.1 Introduction; 4.2 LeNet-5; 4.3 AlexNet; 4.4 ZFNet; 4.5 VGGNet; 4.6 GoogleNet; 4.7 ResNet; 4.8 Densely Connected Convolutional Network (DenseNet); 4.9 Capsule Network; 4.10 Challenges and Future Research Direction; Bibliography; 5 Unsupervised Deep Learning Architectures; 5.1 Introduction 5.2 Restricted Boltzmann Machine (RBM)5.2.1 Variants of Restricted Boltzmann Machine; 5.3 Deep Belief Network; 5.3.1 Variants of Deep Belief Network; 5.4 Autoencoders; 5.4.1 Variations of Auto Encoders; 5.5 Deep Autoencoders; 5.6 Generative Adversarial Networks; 5.7 Challenges and Future Research Direction; Bibliography; 6 Supervised Deep Learning in Face Recognition; 6.1 Introduction; 6.2 Deep Learning Architectures for Face Recognition; 6.2.1 VGG-Face Architecture; 6.2.2 Modified VGG-Face Architecture; 6.3 Performance Comparison of Deep Learning Models for Face Recognition 6.3.1 Performance Comparison with Variation in Facial Expression6.3.2 Performance Comparison on Images with Variation in Illumination Conditions; 6.3.3 Performance Comparison with Variation in Poses; 6.4 Challenges and Future Research Direction; Bibliography; 7 Supervised Deep Learning in Fingerprint Recognition; 7.1 Introduction; 7.2 Fingerprint Features; 7.3 Automatic Fingerprint Identification System (AFIS); 7.3.1 Feature Extraction Stage; 7.3.2 Minutia Matching Stage; 7.4 Deep Learning Architectures for Fingerprint Recognition; 7.4.1 Deep Learning for Fingerprint Segmentation
• Publisher Details: Singapore : Springer
• Publication Date: 2019
• Extent: 1 online resource
• Subjects: 370.15/23; Education -- Data processing; Learning, Psychology of; Motivation in education; EDUCATION / Essays; EDUCATION / Organizations & Institutions; EDUCATION / Reference; Electronic books; Electronic books
• Languages: English
• ISBNs: 9789811367946; 9811367949
• Related ISBNs: 9789811367939; 9811367930
• Notes: Note: Online resource; title from PDF file page (EBSCO, viewed March 21, 2019).
• Access Rights: Legal Deposit; Only available on premises controlled by the deposit library and to one user at any one time; The Legal Deposit Libraries (Non-Print Works) Regulations (UK).
• Access Usage: Restricted: Printing from this resource is governed by The Legal Deposit Libraries (Non-Print Works) Regulations (UK) and UK copyright law currently in force.