An introduction to models, techniques & algorithms for analyzing and making predictions based on networked (graph) data. Prediction tasks include node classification, link prediction, graph classification, etc. Students will learn about theoretical foundations, building on linear algebra and probability concepts. Additionally, they will gain hands-on experience in working with real-world datasets, where nodes represent non-IID observations. Topics include, but are not limited to: node embeddings, graph neural networks and random graph models. Recommended Background: CS 4342 Machine Learning, MA 2071 Matrices and Linear Algebra I, MA 2621 Probability for Applications.
Concentrates on the study of the internals of database management systems. Topics include principles and theories of physical storage management, advanced query languages, query processing and optimization, index structures for relational databases, transaction processing, concurrency control, distributed databases, and database recovery, security, client server and transaction processing systems. Students may be expected to design and implement software components that make up modern database systems.
Learn about neural networks in both theory and practice. Class time will be focused on presenting the mathematics of neural networks and how to train them, as well as discussing seminal/recent papers in the very active research domain of deep learning. Homework will consist of approximately biweekly homework assignments, assigned papers to read, as well as a final project (the last ~5 weeks of the course).
Introduces core methods in Data Science. It covers a broad range of methodologies for working with large and/or high-dimensional data sets to making informed decisions based on real-world data. Core topics include data collection through use cycle, data management of large-scale data, cloud computing, machine learning and deep learning. Students will acquire experience with big data problems through hands-on project susing real-world data sets.
