Machine Learning (ML)
CY-701-
Unit 1Introduction to Machine Learning
introduction to machine learning · scope and limitations of machine learning · regression · probability · statistics · linear algebra for machine learning · convex optimization · data visualization · hypothesis function and testing · data distributions · data preprocessing · data augmentation · normalizing data sets · machine learning models · supervised learning · unsupervised learning
-
Unit 2Deep Learning Fundamentals
linearity vs non‑linearity · activation functions (sigmoid, ReLU, etc.) · weights and bias · loss function · gradient descent · multilayer network · back‑propagation · weight initialization · training and testing of networks · unstable gradient problem · auto‑encoders · batch normalization · dropout · L1 regularization · L2 regularization · momentum · tuning hyper‑parameters
-
Unit 3Convolutional Neural Networks
convolutional neural network (CNN) · flattening · subsampling · padding · stride · convolution layer · pooling layer · loss layer · 1x1 convolution (depthwise) layer · inception network · input channels · transfer learning · one‑shot learning · dimension reduction · implementation of CNN with TensorFlow · implementation of CNN with Keras
-
Unit 4Recurrent Neural Networks and Reinforcement Learning
recurrent neural network (RNN) · long short‑term memory (LSTM) · gated recurrent unit (GRU) · translation models · beam search and beam width · BLEU score · attention model · reinforcement learning (RL) · RL framework · Markov decision process (MDP) · Bellman equations · value iteration · policy iteration · actor‑critic model · Q‑learning · SARSA
-
Unit 5Support Vector Machines and Applications
support vector machines (SVM) · Bayesian learning · application of machine learning in computer vision · application of machine learning in speech processing · application of machine learning in natural language processing · case study: ImageNet competition