Machine Learning using Python: Theory and Application

This course will focus on the theory and implementation of the most common machine learning models. The students will be using the Python programming language in order to use the taught models and apply them to solve real world problems. Through practical assignments, and a project on real world data the students will learn to independently recognize problems that can be solved via machine learning, choose the right machine learning algorithm to use, and evaluate their results correctly. The machine learning models that will be covered throughout the course include: Linear Regression, Logistic Regression, Support Vector Machines, Principal Component Analysis, Neural Networks, Convolutional Neural Networks, Recurrent Neural Networks, and more.

Learning Goals:

• Theory and tricks of the trade for training the most common machine learning models
• Implement machine learning algorithms using Sklearn, Scikit-learn, PyTorch
• Handle and preprocess data using famous Python data science packages NumPy and Pandas
• Learn to visualize and interpret the results of the trained machine learning models
  

Course Structure:

Reading week: January 4th - January 8th, 2021. Flexible, 10 hours preparatory work to be done on-demand.

Online course: January 11th - January 29th, 2021. Estimated session times are Mondays through Fridays from 9 am to 2 pm CET for live lectures and group sessions, etc.

Please note that exact session times will be confirmed once registrations have closed (sessions will be scheduled according to the time zones of the registered course participants).

Should you have any questions regarding the course timetable, please contact us at summeruniversity@tubs.de [2]

Please note this is a full-time, intensive course. Weeks 1-3 will involve approximately 30 hours of workload.

Syllabus:

A detailed syllabus with information on the schedule will be made available to registered participants.

You may find the syllabus useful when discussing with your home university whether the ECTS credits attainable for this course are accepted by them.

Course Components:

-          live lectures

-          independent assignments

-          webinars for assignments discussion and solutions

-          independent project

-          webinars for project discussion/tutoring

-          online test

This course is designed for current university students, working professionals and any individuals with an interest in furthering their knowledge and skills in programming with Python for Data Science and Machine Learning.

Participants from all fields and disciplines are welcome.

Dennis Grinwald is a Research Assistant in the Machine Learning group at Fraunhofer HHI and a final year graduate student in Computer Science, with specialization in machine learning and artificial intelligence. In his research he mainly focuses on improving efficiency of distributed neural networks. Throughout his academic career he has gained notable experience as a lecturer. 

Sascha Lange is a final year graduate student with significant experience in both teaching and research, in the field of computer science. His lecturing experience includes working with the Neural Information Processing Group at the TU Berlin as a research assistant as well as teaching students in the topic of Reinforcement Learning. He is currently undertaking research on risk sensitive Reinforcement Learning, connecting it to meta-learning, to efficiently obtain behavioral models for human-computer-interaction in a limited data regime.

Course fees for Machine learning using Python: Theory and Application are as follows:

Student: 920 Euro

Working professional/Non-student: 1320 Euro

The early bird discount is available for all participants until November 1^st. Registrations open in October 2020.

Please note that students will be required to upload proof of their student status (student card/ enrolment information) during the registration process.