General
Content
Machine learning (ML) is rising in importance as a new approach in
addition to the more traditional statistical methods in data analysis
and predictive modelling in many fields ranging from natural language
processing to materials science. In chemistry, one of the recent efforts
has been to predict complex physical quantities based on molecular and
materials geometry, without the need of computationally demanding
molecular modelling or ab initio quantum chemistry / DFT calculations.
Fast screening of a large amount of structures is needed for developing
chemical compounds and materials solutions for novel and desired
functionalities. The purpose of the course is to give a first
introduction to the vast field of machine learning (basic concepts,
methods and workflow) and survey how ML can be used in chemistry. The
general aim is to convey the exciting potential of ML, which you may
later on meet or employ in your own work.
The course
involves working on exercise problems/assignments, giving a presentation
and writing a home essay (the exact requirements will be fixed in the beginning of the course). There is no final exam.
Course level and prerequisites
This is an introductory course. Basics of numerical data handling and ability to follow code examples (in python) is expected. Learning outcomes
After this course you
- understand the basics of a ML approach and its concepts from problem setting to evaluating the goodness of a model
- can apply ML to solve a simple task
- have an overview how ML is used in contemporary chemistry
- can analyze and evaluate a chemistry-related ML article
Assessment methods
The exact requirements for passing will be fixed in the beginning of the course.
Lectures and computer exercises (22.3. - 28.4.)
Lectures are planned on Wednesdays 12-14 and Thursdays 10-12. Exercises are planned on Fridays 13-15.
First lecture: Wednesday 22.3. in B202b 12:15 - 14:00
For students: Remember to register also in Oodi.