Machine learning and data science

Machine learning and data science (winter term 2021/2022)

Steffen Rulands

Time: Tuesdays, 4:40pm CET

Location:

Online via Zoom (link)

Meeting ID: 854 1370 2254
Passcode: 090867    

Limited physical attendance if regulations allow:

Seminar Room 4
Max Planck Institute for the Physics of Complex Systems
Noethnitzer Str. 38
01187 Dresden

Summary: In recent years, the availability of large sets of data has led to the development of computational methods for their manipulation and interpretation. In this lecture, we will introduce fundamental concepts from data science and machine learning, such as dimensionality reduction, neural networks and deep learning.

Required prior knowledge: Programming

Course materials

Password for PDF files is the last name of the lecturer.

Overview over the lecture - types of machine learning tasks - challenges in machine learning

Lecture notes (PDF)

Introduction to unsupervised learning - density estimation - restricted Boltzmann machines

Lecture notes (PDF)

Source code (Jupyter notebook)

Introduction to clustering - clustering algorithms - clustering validation

Lecture notes (PDF)

Guest lecture by Fabian Rost: Dimensionality reduction - practical example of genome data analysis

Lecture notes (PDF)

Classification - short introduction to statistics - naive Bayes classifier

Lectures notes (PDF)

Decision trees and support vector machines

Lectures notes (PDF)

Regression and regularisation

Lectures notes (PDF)

Biological and artificial neural networks

Lecture notes (PDF)

Statistical physics of deep neural networks

Lecture notes (PDF)

Statistical physics of deep neural networks - Convolutional neural networks - Recurrent neural networks - Generative adversarial neural networks

Lecture notes (PDF)

Reinforcement learning

Lecture notes (PDF)

Research seminar by Anna Poetsch (BIOTEC) on using deep neural networks to predict genome instability in cancer.

Lecture notes and video recording not available.