Overview of course material: Data Analysis and Machine Learning
Morten Hjorth-Jensen [1, 2]
[1] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University, USA
[2] Department of Physics (office FV308), University of Oslo, Norway
The teaching material is produced in various formats for printing and on-screen reading.
Warning.
The PDF files are based on LaTeX and have seldom technical
failures that cannot be easily corrected.
The HTML-based files, called "HTML" and "ipynb" below, apply MathJax
for rendering LaTeX formulas and sometimes this technology gives rise
to unexpected failures (e.g.,
incorrect rendering in a web page despite correct LaTeX syntax in the
formula). Consult the corresponding PDF
files if you find missing or incorrectly rendered
formulas in HTML or ipython notebook files.
Basic introduction to the course with schedule etc
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Introduction to Data Analysis and Machine Learning
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Getting started with Machine Learning with simple Examples
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Review of central linear algebra elements
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Monte Carlo methods and elements of probability theory
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Regression Methods
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Gradient methods and Minimization Algorithms
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Logistic Regression
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Neural Networks
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Convolutional Neural Networks
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Reduction of dimensionality
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Decision Trees and Random Forests
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Support Vector Machines
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Unsupervised Learning, Boltzmann Machines
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Recurrent Neural Networks
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Autoencoders
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Reinforcement Learning
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Solving ordinary and Partial Differential Equations and Eigenvalue Problems with Neural Networks
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Elements of Bayesian theory and Bayesian Neural Networks
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Summary
- LaTeX PDF:
- HTML:
- Jupyter notebook:
Projects and Exercises Fall 2019
First homework set, week 35 (Available August 20)
Second homework set, week 36 (Available August 26)
Project 1, Deadline October 7 (available August 26)
Project 2, Deadline November 8 (available October 2)
Project 3, Deadline December 15 (available November 4)
Course content
Probability theory and statistical methods play a central role in science. Nowadays we are
surrounded by huge amounts of data. For example, there are about one trillion web pages; more than one
hour of video is uploaded to YouTube every second, amounting to 10 years of content every
day; the genomes of 1000s of people, each of which has a length of \( 3.8\times 10^9 \) base pairs, have
been sequenced by various labs and so on.
This deluge of data calls for automated methods of data analysis,
which is exactly what machine
learning provides. In this course the approach is to define machine learning as a set of methods that can
automatically detect patterns in data, and then use the uncovered patterns to predict future
data, or to perform other kinds of decision making under uncertainty. Since many of these problems can be studied using
tools of probability theory, the aim of this course is to expose you to central methods in probability theory linked with machine learning.
This course covers thus topics like Monte Carlo methods and Markov chains, Bayesian statistics, error estimates, various linear methods, optimization of data and error analysis and central algorithms in machine learning.
The course has several numerical projects and numerical exercises that are meant to illustrate the theory.
Learning outcomes
The course introduces a variety of central algorithms and methods
essential for studies of data analysis and machine learning. The course is project based and through the various projects, normally three, the students will be exposed to fundamental research problems in these fields, with the aim to reproduce state of the art scientific results. The students will learn to develop and structure large codes for studying these systems, get acquainted with computing facilities and learn to handle large scientific projects. A good scientific and ethical conduct is emphasized throughout the course. More specifically, after this course you will
- Learn about basic data analysis, Bayesian statistics, Monte Carlo methods, data optimization and machine learning;
- Be capable of extending the acquired knowledge to other systems and cases;
- Have an understanding of central algorithms used in data analysis and machine learning;
- Have a basic knowledge of Bayesian statistics and learning and common distributions;
- Gain knowledge of central aspects of Monte Carlo methods, Markov chains, Gibbs samplers and their possible applications, from numerical integration to simulation of stock markets;
- Understand linear methods for regression and classification;
- Learn about neural network, genetic algorithms and Boltzmann machines;
- Work on numerical projects to illustrate the theory. The projects play a central role and students are expected to know modern programming languages like Python or C++.
Prerequisites
Basic knowledge in programming and numerics. Required courses are the equivalents to the University of Oslo mathematics courses MAT1100, MAT1110, MAT1120 and at least one of the corresponding computing and programming courses INF1000/INF1110 or MAT-INF1100/MAT-INF1100L/BIOS1100/KJM-INF1xxx.
The course has two central parts
- Statistical analysis and optimization of data
- Machine learning
Statistical analysis and optimization of data
The following topics will be covered
- Basic concepts, expectation values, variance, covariance, correlation functions and errors;
- Review of linear algebra methods;
- Splines and conjugate gradient methods for data optimization;
- Simpler models, binomial distribution, the Poisson distribution, simple and multivariate normal distributions;
- Central elements of Bayesian statistics and modeling;
- Monte Carlo methods, Markov chains, Metropolis-Hastings algorithm, ergodicity;
- Linear methods for regression and classification;
- Estimation of errors using blocking, bootstrapping and jackknife methods;
- Practical optimization using Singular-value decomposition and least squares for parameterizing data.
Machine learning
The following topics will be covered
- Gaussian and Dirichlet processes;
- Decision trees and nearest neighbors;
- Support vector machines;
- Boltzmann machines;
- Neural networks;
- Genetic algorithms.
All the above topics will be supported by examples, hands-on exercises and project work.
Recommended textbooks
General learning book on statistical analysis:
- Christian Robert and George Casella, Monte Carlo Statistical Methods, Springer
- Peter Hoff, A first course in Bayesian statistical models, Springer
General Machine Learning Books:
- Kevin Murphy, Machine Learning: A Probabilistic Perspective, MIT Press
- Christopher M. Bishop, Pattern Recognition and Machine Learning, Springer
- David J.C. MacKay, Information Theory, Inference, and Learning Algorithms, Cambridge University Press
- Trevor Hastie, Robert Tibshirani, and Jerome Friedman, The Elements of Statistical Learning, Springer
- David Barber, Bayesian Reasoning and Machine Learning, Cambridge University Press