CS627 : Graphical Models and Structured Learning

• Class Timings: Wednesdays and Fridays: 9:30 AM to 11:00 AM
• Venue: SIC 301, KReSIT.
• Number of credits: Six
• Instructor's office hours: Tuesdays and Fridays: 3:30 to 4:30 PM.

Probabilistic graphical models provide a framework for probabilistic reasoning about several interdependent variables represented as a graph of dependencies. This is a classical topic with roots in statistical physics. In recent years, spurred by several applications in biology, image processing, vision, unstructured data integration and code design, the topic has received renewed interest in the machine learning and data mining communities. The course will attempt to provide a foundational overview of the field while also covering the recent research literature and various applications. To know more, check out Kevin Murphy's excellent introduction to the topic.

The course is targeted for CS/IT research scholars, MTechs, third/final year BTechs and dual degree students who have already taken a first course on machine learning, data mining or statistical data analysis. Think of this as a specialized elective to be taken by students with research interest in the area.