| January 14th, 2021 10:58 AM | |
| vikash | Sathyabama Institute of Science and Technology BE CSE SCSA3015 Deep Learning Syllabus Sathyabama Institute of Science and Technology BE CSE SCSA3015 Deep Learning Syllabus SATHYABAMA INSTITUTE OF SCIENCE AND TECHNOLOGY SCHOOL OF COMPUTING SCSA3015 DEEP LEARNING L T P Credits Total Marks 3 0 0 3 100 UNIT 1 INTRODUCTION 9 Hrs. Introduction to machine learning- Linear models (SVMs and Perceptrons, logistic regression)- Intro to Neural Nets: What a shallow network computes- Training a network: loss functions, back propagation and stochastic gradient descent- Neural networks as universal function approximates. UNIT 2 DEEP NETWORK 9 Hrs. History of Deep Learning- A Probabilistic Theory of Deep Learning- Backpropagation and regularization, batch normalization- VC Dimension and Neural Nets-Deep Vs Shallow Networks- Convolutional Networks- Generative Adversarial Networks (GAN), Semi-supervised Learning. UNIT 3 DIMENTIONALITY REDUCTION 9 Hrs. Linear (PCA, LDA) and manifolds, metric learning - Auto encoders and dimensionality reduction in networks - Introduction to Convnet - Architectures – AlexNet, VGG, Inception, ResNet - Training a Convnet: weights initialization, batch normalization, hyper parameter optimization. UNIT 4 OPTIMIZATION AND GENERALIZATION 9 Hrs. Optimization in deep learning– Non-convex optimization for deep networks- Stochastic Optimization- Generalization in neural networks- Spatial Transformer Networks- Recurrent networks, LSTM - Recurrent Neural Network Language Models- Word-Level RNNs & Deep Reinforcement Learning - Computational & Artificial Neuroscience UNIT 5 CASE STUDY AND APPLICATIONS 9 Hrs. Imagenet- Detection-Audio WaveNet-Natural Language Processing Word2Vec - Joint Detection- BioInformatics- Face Recognition- Scene Understanding- Gathering Image Captions. Max.45 Hrs. COURSES OUTCOMES On completion of the course, student will be able to CO1 - Understand basics of deep learning. CO2 - Implement various deep learning models. CO3 - Realign high dimensional data using reduction techniques. CO4 - Analyze optimization and generalization in deep learning. CO5 - Explore the deep learning applications. CO6 - Design and creation of data models. TEXT / REFERENCE BOOKS 1. Cosma Rohilla Shalizi, Advanced Data Analysis from an Elementary Point of View, 2015. 2. Deng and Yu, Deep Learning: Methods and Applications, Now Publishers, 2013. 3. Ian Good fellow, Yoshua Bengio, Aaron Courville, Deep Learning, MIT Press, 2016. 4. Michael Nielsen, Neural Networks and Deep Learning, Determination Press, 2015. END SEMESTER EXAMINATION QUESTION PAPER PATTERN Max. Marks : 100 Exam Duration : 3 Hrs. PART A : 10 Questions of 2 marks each-No choice 20 Marks PART B : 2 Questions from each unit with internal choice, each carrying 16 marks 80 Marks |
