E1 245o Statistical Inference for Engineers and Data Scientists 3:1 (January 2023)
Course Instructor: Sundeep Prabhakar Chepuri, ECE
Course description:
The main goal of the Statistical Inference for Engineers and Data Scientists course is to cover the major domains of statistical inference, namely, estimation, detection and learning, which include the many mathematical tools that engineers and statisticians use to draw inference from imperfect or incomplete data. The first part of the course develops statistical parameter estimation methods to extract information from data in noise. The second part of this course is about the application of statistical hypothesis testing to the detection of data in noise.
Syllabus
Minimum variance unbiased estimators, maximum likelihood theory, the Cramer-Rao bound, best linear unbiased estimators, Bayesian estimation techniques, the Wiener and Kalman filters, generative models, regression and regularization, binary and multiple hypothesis testing, Neyman-Pearson detector, Bayes detector, and composite hypothesis testing.
Course level: This is a 3:1 credit core course for the M.Tech (Online) ECE program and relevant for AI and DSBA programs as well.
Textbooks / References
- Fundamentals of Statistical Signal Processing, Volume I: Estimation Theory, S.M. Kay, Prentice Hall 1993, ISBN-13: 978-0133457117.
- Fundamentals of Statistical Signal Processing, Volume II: Detection Theory, S.M. Kay, Prentice 1993, ISBN-13: 978-0135041352.
Prerequisites: Matrix theory and Random processes.
Grading and other requirements:
The weightage will be as follows. Three assignments 10% each (i.e., 30 % in total), mid-term exam 20%, final project 30%, and final examination 20%. Three assignments and the mid-term shall be treated as sessional assessment, while the final project and final examination shall be treated as the final assessment.