CSE 588: Large-Scale Machine Learning
Course Instructor: Dr. Mehrdad Mahdavi
Topics covered: Probably approximately correct (PAC), Empirical Risk Minimization (ERM), Linear Algebra, Matrix Computation, Convex Analysis, Non-Convex Analysis, Gradient Descent, Stochastic Gradient Descent, Accelerating Gradient Descent, Projection Free Optimization, Non-Convex Optimization, Landscape Analysis, Fast training of DNNs: learning rate schedules and initialization, Accelerated inference for DNNs: model compression, quantization, distillation, compression via pruning, Hyperparameter optimization: naive search methods & Bayesian optimization, Hyperparameter optimization: online learning and gradient-based methods, Concentration inequalities: basics, Concentration inequalities: sub-Gaussian and sub-Exponential random variables, Random Fourier Features, Randomized Linear Algebra.
IST 557: Data Mining I - Techniques and Applications
Course Instructor: Dr. Justin Silverman
Topics studied: Linear Regression, Bayesian Regression, Penalized Regression, Non Gaussian Regression, Gaussian Process Regression, SVM, Dimensionality Reduction Methods, Naive Bayes, Decision Trees, Random Forests, Ensemble Methods, K-Means Clustering, Mixture Models, Topic Models, Recommender Systems, Anomaly Detection, Bayesian Decision Theory, Bayesian Optimization, Bayesian Inference.
IST 558: Data Mining II
Course Instructor: Dr. Justin Silverman
Topics studied: Bayesian Statistics, MCMC, Probabilistic Programming, Regression in STAN, Missing Data, Compositional Data Analysis, Functional Data Analysis, Dynamic Linear Models, Dirichlet Process, Variational Inference, Variational Autoencoders, Generative Adversarial Networks, Boosting
IST 597 (Special Topics): Foundations of Deep Learning
Course Instructor: Dr. C. Lee Giles
Topics studied: Deep Feedforward Networks, Regularization, Optimization for Training Deep Models, CNN, RNN, LSTM, Autoencoders, Generative Adversarial Networks, Transformers, Neural Architectures with memory, Neural Turing Machines, RNN - Automata, Memory, and Grammar, Science inspired Deep Learning, Physics-Informed Neural Networks.
IST 597 (Special Topics) : Fairness, Incentives, and Mechanism Design
Course Instructor: Dr. Hadi Hosseini
Topics studied: Voting, Cake Cutting, Indivisible goods, Random Assignment, Matching
STAT 500: Applied Statistics
Course Instructor: Dr. Priyangi Bulathsinhala
Topics studied: Numerical Descriptive Statistics, Minitab, Basic probability rules and properties, Conditional Probability and Independence, Discrete Random Variables, Binomial Distribution, Normal Distribution, Central Limit Theorem, Confidence Intervals (C.I) for Mean, Comparing two means using C.I, C.I for proportions, Determining sample size, Hypothesis Testing (H.T) for one mean, H.T for comparing two means, H.T testing about proportions, Chi-squared test for independence, Simple Linear Regression, Comparing two variances, ANOVA.