(Special Topics (Deep Learning
- Introduction to Neural Networks, Origin of Neural Networks, Perceptron, Decision surface, Expressiveness of a perceptron, Activation functions & Perceptron Learning, Perceptron Learning Summary
- Multilayer Perceptron, Backpropagation, gradient descent on the error surface, Gradient descent Learning, Batch/Stochastic/ Minibatch, Issues with ANN
- Why are deep Neural Networks hard to train
- Deep Learning, Introduction to Convolutional Networks, Recent Progress to deep Neural Nets, on the future of Neural Networks,
- Convolutional Neural Networks, Basic Structure of a Convolutional Neural Network, Filter for Convolutional operation, Convolutional operation, Effects of convolution, need for padding, Types of padding, Strided Convolution, Max pooling, Relu. Fully connected layers, Stuck at the endو interleaving between layers.
- Presenting a Conference Paper from IJCNN conference