Keep your paper neat. Office: Centergy One, Room 5212 ECE 6254 Statistical Machine Learning Spring 2017 Mark A. Davenport Georgia Institute of Technology School of Electrical and Computer Engineering. Email: mdav (at) gatech (dot) edu Folding the corners, etc., does not count. ECE 6254 - Statistical Machine Learning Spring 2019. Accordingly, the following are guidelines that are aimed at improving the quality of homework submissions, facilitating learning, and helping the grader to quickly evaluate your submission. If you’re unsure about taking the class, the self-assessment is here to help! This course will provide an introduction to the theory of statistical learning and practical machine learning algorithms with applications in signal processing and data analysis. ECE 6254 - Statistical Machine Learning Watch Piazza for information about assignments. ECE 6254: Statistical Machine Learning Spring 2017. RESOURCES. Summer 2014, ECE 3770, Intro to Probability and Statistics for ECEs. We will cover the tools from convex optimization that we need as part of the course, but if you want to know more, this is a great resource targeted towards electrical engineers. The material for this course will come from several different sources. Background image credit: www.freepik.com. Exam information is on the Schedule page. We will approach these problems from the perspective of statistical inference. ECE 6254: Statistical Machine Learning Spring 2017. Finally, many of the homework assignments and the course projects will require the use of Python. Mark Davenport If you use a cell-phone camera to scan make sure that you use a scanner app that crops to the paper, removes shadows, and generally creates a high quality scan that is easily printable. You may also use a cover sheet. Textbooks . Do not use paper with rough edges from being torn out of a notebook. It should be easy to identify the answer to each problem at a glance. We will also be using the language of linear algebra to describe the algorithms and carry out any analysis, so you should be familiar with concepts such as norms, inner products, orthogonality, linear independence, eigenvalues/vectors, eigenvalue decompositions, etc. Mark Davenport Email: mdav (at) gatech (dot) edu Office: Centergy One, Room 5212 Phone: (404)894-2881 Office Hours: TBD Description. A first model of learning: Concentration inequalities and generalization bounds, The Bayes classifier and nearest neighbors classifiers, Plugin methods I: NaÃ¯ve Bayes and linear discriminant analysis, More linear classifiers: The perceptron algorithm and maximum margin hyperplanes, Theory of generalization: Dichotomies, the growth function, shattering, and break points, Theory of generalization 2: The Vapnik-Chervonenkis generalization bound, Regression, least squares, and Tikhonov regularization, The LASSO, robust regression, kernel regression, and regularization in classification, Overfitting and the bias-variance tradeoff, Dimensionality reduction, feature selection, and principal component analysis, Kernel density estimation and k-means clustering, Gaussian mixture models and expectation maximization, Spectral clustering, density-based clustering, and hierarchical clustering. as well as the basics of multivariable calculus such as partial derivatives, gradients, and the chain rule. 01/28/2019 Problem set 1 available on Canvas. ECE6254 at Georgia Institute of Technology for Summer 2020 on Piazza, a free Q&A platform for students and instructors. You are encouraged to study the solutions for every problem and compare with your own. (, Some linear algebra and other problems from ECE 6250, Introduction to Statistical Learning Theory, Python for Data Analysis, McKinney, 2012. Machine Learning: A Probabilistic Perspective by Murphy (2012). ECE 6254 - Statistical Machine Learning.
ECE6254 Course Syllabus ECE6254 Statistical Machine Learning (3-0-3) Prerequisites ECE 4270 Corequisites None Catalog Description An introduction to the theory of statistical learning and practical machine learning algorithms with applications in signal processing and data analysis.
Statistical machine learning • How can we – learn effective models from data? Watch Piazza for information about assignments. (Hardback available for ~$90 on amazon). There are many other books and journal papers of interest which will be listed in the resources section of the course web site. Introduction; Introductory supervised learning. Overview; Course Notes; Assignments; Resources; Papers. Practice brevity while maintaining completeness. Introduction to Statistical Learning Theory, Bousquet, Boucheron, and Lugosi, 2004. If you have had courses on these topics as an undergraduate (or more recently) you should be able to fill in any gaps in your understanding as the semester progresses. This course will provide an introduction to the theory of statistical learning and practical machine learning algorithms with applications in … We will consider a variety of applications, including classification, prediction, regression, clustering, modeling, and data exploration/visiualization. The rest of the material will come from supplementary lecture notes. : An interesting podcast about machine learning, Mathematical Methods and Algorithms for Signal Processing, , Polya, 1945. Phone: (404)894-2881 It is important to get to the heart of a problem and not cloud your submission with fluff. In contrast to most traditional approaches to statistical inference and signal processing, in this course we will focus on how to learn effective models from data and how to apply these models to practical signal processing problems.
Put your name in the upper right corner along with the date and class (ECE 6254). Historical grade point average (GPA) for professors and courses offered at the Georgia Institute of Technology Use sufficient paper. The application of learned principles, and practice, are essential to learning new material. Overview; Course Notes; Assignments; Resources; Course overview. Jan 07, 2020: Syllabus Overview - The Supervised Learning Problem: Jan 09, 2020: Why supervised learning may work: Jan 14, 2020: Why supervised learning may work All answers must be clearly indicated (boxed if appropriate).
Classic introduction to mathematical problem solving. There is a change in the submission process: the assignments are to be scanned and uploaded to the. This information should appear on every page of your submission. Write on one side of your paper. (where required, code may be in a separate file), due Monday, February 5 in class for Atlanta students, due Monday, February 12 for remote students, due Wednesday, February 14 in class for Atlanta students, due Monday, February 19 for remote students (submit via the, (updated to make problem 3.1 clearer on 2/16), due Monday, February 26 in class for Atlanta students, due Monday, March 5 for remote students (submit via the, due Wednesday, March 28 in class for Atlanta students, due Wednesday, April 4 for remote students (submit via the Assignments tab in t-square), due Wednesday, April 18 in class for Atlanta students, due Wednesday, April 25 for remote students (submit via the Assignments tab in t-square).