Neural networks are one of the most popular machine learning algorithm today, achieving impressive performance on a large variety of tasks where traditional ml models are not good. Topics to be covered: Foundation • Neural Network analogy to Human Brain • NN Representation, Activation functions, Backpropagation, • Training (weight optimization) using backpropagation • Gradient descent, Stochastic gradient descent • Hyperparameter tuning • Deep neural networks: preventing overfitting Convolutional neural networks • convolutional neural networks Architecture and Layers • Image Classification • Object detection • One stage methods: YOLO and SSD • Two stage methods: Faster R-CNN • Facial recognition Recurrent neural networks • RNN Model • Vanishing gradient problem • Gated recurrent units: Introducing intentional memory • Long short term memory networks: Learning what to remember and what to forget Transfer learning • Image recognition • Natural language processing Relevant Libraries: • Python — Keras, Tensorflow, caffe, mxnet, theano, deeplearning4j • R — h2o, neuralnet, nnet, tensorflow, rcppDL, MXNetR