Chapter 1: Machine Learning Basics and Mathematical Foundation for Deep Learning <div>Chapter Goal: Introduce Machine Learning basics and Mathematical Foundations that are associated with Deep Learning </div><div>No of pages 70-90</div><div>Sub-Topics</div><div>1. Linear Algebra basics.</div><div>2. Numerical Stability and Conditioning.</div><div>3. Probability.</div><div>4. Different types of cost functions and introduction to least squares and maximum likelihood methods.</div><div>5. Convex and Non-convex function </div><div>6. Optimization Techniques such as Gradient Descent and Stochastic Gradient Descent as well as Constrained Optimization problems.</div><div>7. Regularization and Early stopping</div><div>8. Auto Differentiators and Symbolic Differentiators.</div><div><br/></div><div>Chapter 2: Introduction to Deep Learning Concepts and TensorFlow </div><div>Chapter Goal: Introduce Deep Learning concepts and its comparison with previous Neural Netwo</div>rks. Reasons for its success and computational efficiency and a start to TensorFlow Development.<div>No of pages 60-70</div><div>Sub -Topics </div><div>1. Previous Neural Networks and their shortcomings </div><div>2. Introduction to Deep Learning Framework and its advantages.</div><div>3. Why TensorFlow for Deep Learning and its comparison with other Deep Learning Frameworks like Theano, Caffe, Torch, etc.</div><div>4. Hands on in TensorFlow development environment and introduction to Dynamic Computation graphs. </div><div>5. Linear and Logistic regression in a TensorFlow environment</div><div>6. Feed forward networks through TensorFlow.</div><div>7. Leveraging GPUs for Computational efficiency.</div><div><br/></div><div>Chapter 3: Image and Audio Processing in TensorFlow through Convolutional Neural Networks </div><div>Chapter Goal: Learn to process image and audio data to solve classification, clustering, and recommendation problems using Convolutional </div>Neural Network. <div>No of pages: 70-80</div><div>Sub - Topics: </div><div>1. Convolution and Image processing through Convolution.</div><div>2. Different Kinds of Image processing filters like Guassian Filter, Sobel Filter, Canny’s edge detection filter.</div><div>3. Different Layers of Convolutional Neural Network – Convolution layer, Pooling Layers, activation layers using RELUs, Dropout layers and fully connected layer. Intuition of features learned in Different layers. Concepts of strides, padding and kernels.</div><div>4. Solving image classification, clustering and recommendation problems through Convolutional Neural network.</div><div>5. Feature transfer in Convolutional Neural Network.</div><div>6. Audio classification problems through Convolutional Neural networks.</div><div><br/></div><div>Chapter 4: Restricted Boltzmann Deep Learning Architectures through TensorFlow for Various Problems</div><div>Chapter Goal: Leverage Restricted Boltzmann Machines (R</div>BMs) for solving Recommendation problems, weight initialization in Deep Learning Networks and for Layer by Layer training of Deep Neural Networks.<div>No of pages:50-60</div><div>Sub - Topics: </div><div>1. Introduction to Restricted Boltzmann Machines (RBMs) and its architecture.</div><div>2. Using RBMs to build Recommendation engines.</div><div>3. RBMs for smart weight initialization of Deep Learning Networks.</div><div>4. Train complex deep learning networks layer by layer (one layer at a time) through RBMs</div><div><br/></div><div><br/></div><div><br/></div><div>Chapter 5: Deep Learning for Natural Language Processing through TensorFlow </div><div>Chapter Goal: Leverage TensorFlow Deep learning capabilities for Natural Language processing </div><div>No of pages: 50-60</div><div>1. Text processing basics such as Word2Vec Representation, Semantic and Syntactic Analysis. </div><div>2. Recurrent Neural network(RNNs) for language modelling through TensorFl</div>ow<div>3. Backpropagation through time and problems of Vanishing and Exploding gradients.</div><div>4. Gradient Clipping and LSTM (Long Short-Term Memory) to overcome Exploding and Vanishing gradient problems.</div><div>5. Applications of RNN in generating sequences and words.</div><div><br/></div><div>Chapter 6: Unsupervised Learning in TensorFlow through Autoencoders </div><div>Chapter Goal: Leverage Autoencoders for doing Unsupervised Learning </div><div>No of pages: 30-40</div><div>1. Data Compression through Autoencoders.</div><div>2. Feature Learning through Auto Encoders.</div><div>3. A comparison of feature learning through PCA and Stacked Auto Encoders.</div><div><br/></div>
