Petnica Summer Institute

Summer School on Machine Learning for Scientific Research
1 August - 10 August, 2025

We are pleased to announce the Machine Learning for Scientific Research, a summer school aimed at the role of machine learning techniques in different scientific disciplines.

The Aim

The summer school aims to equip participants with an in-depth knowledge of machine learning approaches and to inspire their application in scientific research. To this end, we will offer intensive and thorough courses on fundamental and applied facets of machine learning. Namely, we aim to present how and why techniques work in a specific scientific context.

This way, we want to foster a research mindset: lectures will emphasize methodological principles of machine learning, while accompanying hands-on sessions and independent project work will expose participants to real-world scientific challenges.

For Whom

The school is primarily intended for early graduate (MSc or junior PhD level) or senior undergraduate students. More senior PhD students looking to expand their machine learning skills or build a solid foundation are also encouraged to apply. A background in basic mathematics and familiarity with scientific research is expected. Prior exposure to machine learning is useful but not mandatory. Students from Southeast Europe and the Balkans region are especially encouraged to apply.

By Whom

Lectures will be delivered by a group of early- and mid-career researchers from international institutions, specializing in various applications of machine learning in science and engineering.

Topics

• Deep Learning
• Transformers
• Reinforcement Learning
• Self-Supervised Learning
• Physics-informed neural networks
• Bayesian Networks
• Normalizing Flows

With the applications including, but not limited to: Computer Vision, Signal Processing, Hydrodynamics, Genetics, Finance, Cybersecurity, Linguistics, Astrophysics / Cosmology, etc.