Make sense of complex models and data with modern inference techniques. This course provides a practical introduction to simulation-based inference and its growing role in scientific research.
This course is offered by Helmholtz AI in cooperation with HIDA
Simulation-Based Inference (SBI) is rapidly emerging as a powerful paradigm for scientific discovery, enabling parameter estimation, uncertainty quantification, and model selection in complex systems. Moreover, it can provide a tool to approach inverse problems. This method has rapidly gained widespread attention in the scientific community as it combines statistically sound inference with the large potentials of machine learning.
This course offers the fundamentals of using SBI in practice. We will dive into the underlying theory and embark on a journey to explore the sbi toolbox with practical examples.
Learning goals
The course aspires to provide a mental model of which tasks are suited for SBI and how to approach them. We will provide practical first encounters with the sbi toolbox using jupyter notebooks. We will also touch upon the fact, how to check the validity of each trained density estimator and much more.
Course date
Register now:
For more information on how to register, please follow the link on the course date.
Prerequisites
If you want to enroll in this course, we expect you to bring along knowledge of the Python language as taught the courses "Kickstart Python" and “First Steps in Python”. We also assume that you have acquired the use of pytorch for data science applications and machine learning, e.g. as taught in the “Introduction to Deep Learning” course.
Target group
This course targets researchers interested in performing statistical inference using simulation-based inference.
This course is free of charge.
