![](docs/Kumar.jpg)
Course Description
First level graduate course on Machine Learning. Emphasis on fundamental theory for learning supervised classification/regression models covers Bayes classifier, maximum-likelihood estimation, least squares, Probably Approximately Correct Learning, empirical risk minimization, Vapnik-Chervonenkis Dimension, computational learning, structural risk minimization, regularization, cross-validation, acyclic feedforward networks, completeness of neural networks, backpropagation algorithm, gradient descent, stochastic gradient descent, Convolutional Neural networks, Auto-encoders, Generative Adversarial Networks, support vector machines, kernel-based methods, learning from experts, boosting, Gaussian process- based learning, word embeddings, recurrent neural networks, decision trees, random forests, nearest neighbor classification, seq2seq models, NLP, attention mechanisms, BERT, GPT.
Pre-requisites
- Basic Probability Theory
- Linear Algebra fundamentals
- Calculus
- Quick overview of Convex Optimization
- Hands on Python
- Use of high performance computing resources: HPRC/Google Colab/Department Clusters
Course Pre-requisites
ECEN 303 or MATH 411 or STAT 614 or STAT 615 or by Professor approval.
References
Shai Shalev-Shwartz and Shai Ben-David, Understanding Machine Learning: From Theory to Algorithms, Cambridge University Press, 2014.