Syllabus : Machine Learning: Fundamentals; Neural Network: Perceptrons, Back Propagation, Over-fitting,
Regularization. Deep Networks: Definition, Motivation, Applications; Principal Component Analysis;
Restricted Boltzmann Machine; Sparse Auto-encoder; Deep Belief Net; Hidden Markov Model. Convolution
Neural Network (CNN): Basic architecture, Activation functions, Pooling, Handling vanishing gradient
problem, Dropout, Greedy Layer-wise Pre-training, Weight initialization methods, Batch Normalization;
Different CNN Models: Alex Net, VGG Net, Google Net, Res Net, Dense Net, MIL, Highway Network,
Fractal Network, Siamese Net; Graphical Model: Bayes Net, Variational Auto-encoders. Sequence
Learning: 1D CNN, Recurrent Neural Network (RNN), Gated RNN, Long short-term memory (LSTM).
Generative Modeling: Generative adversarial network. Zero Shot Learning. Applications. |
References : 1. Ian Goodfellow, Yoshua Bengio and Aaron Courville, Deep Learning, MIT Press, 2016
2. Michael A. Nielsen, Neural Networks and Deep Learning , Determination Press, 2015
3. Yoshua Bengio, Learning Deep Architectures for AI, now Publishers Inc., 2009 |