IST Austria Courses
IST Austria logo

Methods of Data Analysis

Instructor: Gasper Tkacik

Teaching Assistant: Katarina Bodova

Description

 

This course introduces a variety of data analysis and simulation methods. It is organized around week-long modules, each covering one method and consisting of 2 lectures, a recitation, and an extensive problem set. The aim is for the students to both understand the method and try it out on real or simulated data. This is a hands-on course that should provide useful practical experience. The students may find the background of DSSC Track Core course helpful, but it is not required.


Prerequisites

Primarily DSSC students but open to any student with: (i) sufficient math background (linear algebra, basic calculus; typically at the level of intro Physics/CS/Engineering/Math undergrads); (ii) sufficient coding capability (working knowledge of a language that supports numerical computation,  e.g., Matlab, Mathematica, C, Python, etc).

 


Credits

3 ECTS 


Final Grade

100% from problem set assignments, 6 in total.

 

Schedule (subject to change)

Date Topic Location Other
Oct 11 Random numbers, Monte Carlo integration Mondi 1  
Oct 13 Stochastic Simulation Algorithm Mondi 1  
Oct 18 Metropolis Monte Carlo and Ising systems Mondi 1  
Oct 20 Entropic sampling Seminar room LBE  
Oct 25 Basics of probabilistic inference Mondi 1  
Oct 27 Probabilistic inference, regularization Mondi 1  
Nov 3 Generalized Linear Models Mondi 1  
Nov 8 Introduction to information theoretic quantities Mondi 1  
Nov 10 Entropy and information estimation Lakeside view room  
Nov 15 Kernel density estimation, Gaussian Mixture Models Mondi 1  
Nov 17 Maximum entropy models Mondi 1  
Nov 22 Bayesian linear regression, feature space Mondi 1  
Nov 24 Gaussian Processes for regression Mondi 1  

Homeworks

Week Due Date
See Week 1 Lecture notes below (Random numbers, SSA) Oct 27
See Week 2 Lecture notes below (Metropolis MC) Nov 3
See Week 3 Lecture notes below (Probabilistic Inference) Nov 10
See Week 4 Lecture notes below (Information theoretic quantities) Nov 24
See Week 5 Lecture notes below (Learning probability distributions) Dec 1

Additional Downloads

Week 1 Lecture notes and Homework
Week 1 Selected chapters from Numerical Recipes

Week 2 Lecture notes and Homework

Week 3 Lecture notes and Homework
Week 3 Required data files (MDA3_spikes.mat MDA3_stim.mat, same data in text format)

Week 4 Lecture notes and Homework
Week 4 Required data files (MDA4_data.zip)

Week 5 Lecture notes and Homework
Week 5 Required data files (MDA5_data.zip)

Week 6 Lecture Notes