2022-08-16 - 2022-08-19 Mathematical Methods for Data Science and Machine Learning
Get an insight into the mathematics behind machine learning.
Reiter
About this training
Applied Mathematics and Machine Learning
A bridge between the pure theoretical knowledge and practice opens up a new dimension of thinking. On one hand you will see how can you develop new mathematical intuition by putting your theoretical question into the proper practical context. On the other hand you will experience how mathematical skills support the understanding of practical solutions from machine learning.
Machine learning has become a universally applicable beloved tool in the recent years. Parallelly to that an increasing number of people are developing curiosity for it. There are available already many handy packages for machine learning. However to benefit the most of it (to be able to interpret the results and to get out the best from a model) good mathematical understanding of the background machinery can make a big difference.
The aim of this module is to remodel your mathematical knowledge by changing the focus from the abstract mathematics itself towards mathematics with a specific purpose and equipping you for a deeper dive into machine learning.
The material of this course will be organised in Jupyter notebooks and the used programming language will be Python.
Reference books:
A bridge between the pure theoretical knowledge and practice opens up a new dimension of thinking. On one hand you will see how can you develop new mathematical intuition by putting your theoretical question into the proper practical context. On the other hand you will experience how mathematical skills support the understanding of practical solutions from machine learning.
Machine learning has become a universally applicable beloved tool in the recent years. Parallelly to that an increasing number of people are developing curiosity for it. There are available already many handy packages for machine learning. However to benefit the most of it (to be able to interpret the results and to get out the best from a model) good mathematical understanding of the background machinery can make a big difference.
The aim of this module is to remodel your mathematical knowledge by changing the focus from the abstract mathematics itself towards mathematics with a specific purpose and equipping you for a deeper dive into machine learning.
The material of this course will be organised in Jupyter notebooks and the used programming language will be Python.
Reference books:
- Ian Goodfellow, Yoshua Bengio, Aaron Courville, Deep Learning (Adaptive Computation and Machine Learning series), 2016
- Andriy Burkov, The Hundred-Page Machine Learning Book, 2019
- Aurélien Géron, Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent System, 2019
Learning outcomes - participants can/know
- Linear algebra
- vector operation and matrix operations,
- projections,
- eigenvalues, eigenvectors,
- Calculus
- extreme values of functions,
- vector-valued functions,
- (directional) gradient of functions,
- approximation of functions,
- chain law,
- backpropagation,
- gradient descent method.
- Regression models
- linear,
- non-linear,
- logistical.
- PCA (principal component analysis)
- Statistics
- combinatorics, probability rules & axioms,
- bayes’ theorem,
- random variables, variance and expectation, conditional and joint distributions, standard distributions (Bernoulli, binomial, multinomial, uniform and Gaussian),
- maximum likelihood estimation (MLE).
Target group
- Students of the CAS on Applied Data Science and of the CAS on Machine Learning.
Prerequisites
- Participants must bring own laptops.
- No programming experience required.
- Graduate maths knowledge is required.
Methods
- The training alternates between short theoretical introductions and practical exercises.
Certificate
- A certificate will be delivered to participants who have attended the whole module.
- Participants attending the module are offered to prepare a project (30 hours). Upon a successful presentation 2 ECTS credit points are given.
Coaches
- The coaches are local or external experts.
Practical information (time, location ...)
Time : 2022-08-16 - 2022-08-19, 09:00 - 12:30 (4 half days)
Location : University of Bern, Mittelstrasse 43, room 324.
Location : University of Bern, Mittelstrasse 43, room 324.
Remote participation: will be available via Zoom. Check the course content (once logged in) to view Zoom link (available closer to the course start date)
Training language: English
Participants : Max 24
Registration : Mandatory
Coaches : Dr. Kinga Sipos (coach), PD Dr. Sigve Haug (responsible)
Prerequisites : Laptop
Certificate : Certificate for full training attendance
Participants : Max 24
Registration : Mandatory
Coaches : Dr. Kinga Sipos (coach), PD Dr. Sigve Haug (responsible)
Prerequisites : Laptop
Certificate : Certificate for full training attendance
About ScITS
The Science IT Support is there to boost your research by supporting you solving computing challenges.
Your code doesn't compile, you need more computing power, more storage, a data mangament plan and
so on - scits.math.unibe.ch.
Your code doesn't compile, you need more computing power, more storage, a data mangament plan and
so on - scits.math.unibe.ch.