Course Credit: 3
Introduction: Intelligent Systems, Expert Systems, Processes, Components and Types of Expert Systems, Neural Network, Fuzzy Logic, Genetics Algorithm. Machine Learning: Definition, General Concepts and Techniques of Machine Learning, Classification of Machine Learning, Supervised, Unsupervised, Reinforcement, Ensemble and Deep Learning, Supervised 51 vs Unsupervised vs Reinforcement Learning, Different Machine Learning Algorithms. Supervised Learning: Definition, General Concept, Problem Solving using Supervised Learning, Learning from a class, Hypothesis, Version Space, Margin, Vapnik-Chervonenkis Dimension, Probably Approximately Correct (PAC) Learning, Noise and Model Complexity, Generalization, Underfitting and Overfitting, Bias-Variance Tradeoff, Triple Tradeoff, Cross Validation, Curse of Dimensionality, Selection of Supervised Learning Algorithms. Linear Regression: Linear Correlation and Regression, Covariance, Correlation Coefficients, Distribution of the Correlation Coefficient, Bivariate and Multivariate Regression Models, Applications, Access the Fit of Regression Model, Coefficient of Determination, Error Matrices, Multiple Linear Regression. Logistic Regression: Logistic Regression, Decision Boundary, Cost Function for Logistic Regression, Gradient Descent Implementation, Multi-class Classification. Regularization: Role of Regularization, How Does Regularization Work?, Modifying the Loss Function, L1 Regularization, L2 Regularization, Elastic Net Regularization . Decision Tree: Definition, General Considerations, Decision Rules, Top-Down Decision Tree Generation, Trees Construction Algorithm (ID3), Information Gain, Attribute Selection, Gini Index, Overfitting and Tree Pruning. Bayesian Decision Theory: Probability and Inference, Classification, Conditional Probability, Bayes’ Theorem, Likelihood, Evidence, Prior, Posterior, Mathematical Examples, Sensitivity and Specificity, Different Losses and Reject, Discriminant Functions, Association Rules, Apriori Algorithm, Maximum Likelihood and Least Squares, Regularized Least Squares. Dimensionality Reduction: Introduction, Subset Selection, Principal Component Analysis, Feature Embedding, Factor Analysis, Singular Value Decomposition and Matrix Factorization, Multidimensional Scaling, Linear Discriminant Analysis, Canonical Correlation Analysis, Independent Component Analysis. Clustering: Definition, Similarity, Euclidean and Non-Euclidian Distance Measures, Partitional and Hierarchical Clustering, Aglomerative Clustering Algorithm, Divisive Clustering, Computing Distance Matrix, Partitional Clustering, K-means Clustering Algorithm, Nearest Neighbor Clustering, The Birch algorithm, Applications of Clustering. Other Clustering Approaches: Density-Based Clustering, Distribution Model-Based Clustering, Fuzzy Clustering, Mean-shift algorithm, DBSCAN Algorithm, Expectation-Maximization Clustering, Agglomerative Hierarchical Algorithm, Affinity Propagation. Support Vector Machine: Introductory Concept, Max-Margin Classifiers, Lagrangian Multipliers, Kernels, Complexity, Linear Classifier, Classifier Margin, Maximum Margin, Linear SVM, Constrained Optimization Problem, Quadratic Programming, Kernel Trick, Overtraining, Practical Example, Performance Measurements, Properties, Applications and Issues with SVM. Artificial Neural Networks: Introduction, History, NN to Solve Problems, Human Biological Neurons, Artificial Neurons, Properties, ANN, Characterizations, Single Layer, Multilayer, Activation Functions, First NN, Perceptron, Training a Perceptron, Problem Domains, Applications of ANN, Practical Examples, ANN 52 vs Expert Systems.
