AI in Physics (M.Sc. Physics – specialization certificate)
Bring cutting-edge AI and machine learning into real physics research — from particle physics to medical imaging and cosmology.
What is the AI in Physics specialization?
The Faculty of Physics at LMU Munich offers an Artificial Intelligence (AI) specialization within the four-semester Master of Science in Physics. As part of the AIM@LMU initiative, you’ll combine solid physics training with advanced machine learning and data-driven methods.
Outcome: You graduate with an LMU Master’s degree in Physics plus an official AI specialization certificate noted by the Examination Office.
Who is it for? Motivated M.Sc. Physics students who want to apply ML/AI to real physics problems across experiment, theory, and simulation.
The certificate combines AI Lab, ML lectures, physics-AI modules, and an AI-focused thesis.
At a glance
- Degree: M.Sc. Physics (with AI specialization certificate)
- Standard duration: 4 semesters
- Language: English (individual courses may vary)
- Start: Winter or Summer semester (course availability varies by term)
- Location: LMU Munich, Faculty of Physics
- Initiative: AIM@LMU
How to earn the AI specialization certificate
- AI Lab – 9 ECTS
- Two ML-focused Master lectures – 12 ECTS (from Computer Science, Mathematics, or Statistics)
- Physics-AI modules – 15 ECTS (lectures/seminars in the Faculty of Physics focused on AI/ML in the physical sciences)
- Practical phases + Master’s thesis at the AI/physics interface (your supervisor confirms eligibility when you register your thesis)
These requirements are in addition to the compulsory components of the Master of Physics course. The Examination Office issues the certificate once all requirements are met.
Information on these regulations can be found here.
Why choose LMU for AI in Physics?
- Breadth of physics + depth in AI: Courses span particle physics, quantum systems, cosmology, complex systems, medical physics, and more.
- Hands-on AI Lab: Design and run experiments such as simulation-based inference for cosmology, AI for particle-physics backgrounds, disease detection from IR spectra, ML for quantum many-body systems, and forecasting chaotic systems.
- Cross-faculty expertise: Learn with instructors from Physics, Computer Science, Statistics, and Mathematics.
- Research-ready skills: Build reproducible workflows, coding competence, and modern ML intuition for scientific discovery.
Courses by semester
Below is the list of courses offered with a relevance for this certificate.
Lectures from Physics:
- Deep Learning for Physicists (Kepesidis, Krausz)
- AI in Physics: When Machine Learning Meets Complex Systems (Räth)
- Inverse problems and machine learning in medical physics (Gianoli)
Seminars:
- Bayesian Inference and Artificial Intelligence (Grün, Heng, Friedrich, Gkouvelis)
Lectures from Computer Science and Statistics:
(note: the main registration for CS courses happens from 1.9.-30.9.2025 on LSF now, late registration for courses without limitations is from 10.10.-30.10.2025)
- Deep Learning for NLP (Aßenmacher)
- Data Mining Algorithmen I (Seidl, Marques Tavares)
- Optimization (Bischl, Schneider)
- Applied Deep Learning (Rügamer, Rezaei)
- Deep Learning and Artificial Intelligence (Schubert)
- Computational Intelligence (Gabor)
- Uncertainty in Artificial Intelligence and Machine Learning (Hüllermeier)
- Intelligent Systems (Bernabé-Moreno)
Lectures from Physics:
- Data Analysis with Machine Learning in Particle Physics (Kuhr, Duckeck, Krug)
- Machine learning for quantum systems (Bohrdt)
Seminars:
- Bayesian Inference and Artificial Intelligence (Grün, Friedrich, Heng, Gkouvelis)
- The Foundations of Artificial Intelligence: Tracing the Pillars of Modern AI (Kepesidis, Gigou, Krausz)
- Non-equilibrium physics of machine learning (Rulands)
- Selected topics from statistical physics and/in AI (Räth)
Lectures from Computer Science and Statistics:
- Computer Vision: Image & Video Understanding (Ommer)
- Machine Learning (Tresp)
- Preference Learning and Ranking (Hüllermeier)
- Deep Learning (Rügamer)
- Natural Computing (Linnhoff-Popien, Gabor)
- Artificial Intelligence for Games (Schubert)
- Interactive Theorem Proving (Blanchette)
- Artificial Intelligence in Interactive Systems (Mayer)
- Supervised Learning (Bischl)
- Applied Deep Learning (Rezaei, Rügamer)
- Infrastructure for Advanced Analytics and Machine Learning (Kranzlmüller)
- Introduction to Foundation Models (also need to apply on moodle, see links in https://ommer-lab.com/teaching) (Ommer)
Labs:
- AI Lab (also need to apply on moodle)
- Experiments:
- X-ray cluster mass determination
- Neural networks for many body quantum physics
- Prediction of students’ physical understanding based on their eye movements during a quantum experiment
- Disease detection using infrared molecular fingerprints
- Simulation-based inference with random fields in cosmology
- Smart Background Simulation for Particle Physics
- Forecasting of chaotic systems
- Differential equations meet automatic differentiation and efficient computation
- Emergent weight morphologies in deep neural networks
- Using quantum computers to solve machine learning tasks
- Bayesian Inference for Spectroscopic Reconstruction
- Experiments:
Lectures from Physics:
- Deep Learning for Physicists (Kepesidis, Krausz)
- AI in Physics: When Machine Learning Meets Complex Systems (Räth)
Seminars:
- Non-equilibrium Physics of artificial intelligence (Rulands)
- Bayesian Inference and Artificial Intelligence (Grün, Heng, Friedrich, Gkouvelis)
Lectures from Computer Science and Statistics:
- Supervised Learning (Bothmann)
- Computational Intelligence (Linnhoff-Popien)
- Data Mining Algorithmen I (Tavares)
- Deep Learning and Artificial Intelligence (Schubert)
- Machine Learning (Tresp)
- Uncertainty in Artificial Intelligence and Machine Learning (Hüllermeier)
- Deep Learning for NLP (Aßenmacher)
- Optimization (Bischl)
- Automated Machine Learning (Feurer)
Lectures from Physics:
- From data to insights (Grün, Friedrich)
- Data Analysis with Machine Learning in Particle Physics (Kuhr, Duckeck, Hartmann)
- AI in Physics: When Machine Learning Meets Complex Systems (Räth)
- Statistical physics of machine learning (Rulands)
Seminars:
- Bayesian Inference and Artificial Intelligence (Grün, Friedrich, Heng, Gkouvelis)
- Causality & Machine Learning (Kepesidis, Gigou, Krausz)
- Current Advances in Machine Learning (Rulands)
Lectures from Computer Science and Statistics:
- Generative AI and Visual Synthesis (Ommer)
- Machine Learning (Tresp)
- Preference Learning and Ranking (Hüllermeier)
- Deep Learning (Rügamer)
- Supervised Learning (Bischl)
- Applied Deep Learning
- Advanced Analytics and Machine Learning (Kranzlmüller, Luckow)
- Natural Computing (Linnhoff-Popien, Gabor)
- Artificial Intelligence for Games (Schubert)
- Interactive Theorem Proving (Blanchette)
- Practical Machine Learning (Mayer)
- Deep Learning for Partial Differential Equations (Bacho)
Labs:
- AI Lab experiments:
- Differential equations meet automatic differentiation and efficient computation
- An analytical test of simulation based inference: using generative models for inference in cosmology
- Hyperspectral imaging and compressed sensing
- Phase transitions in stochastic gases driven by deep neural networks
- Gaussian process regression and Bayesian optimization of Laser-Plasma Accelerators
- Forecasting of chaotic systems
- Prediction of students’ physical understanding based on their eye movements during a quantum experiment
- Smart Background Simulation for Particle Physics
- Disease detection using infrared molecular fingerprints
- X-ray cluster mass determination
- Neural networks for many body quantum physics
Lectures from Physics:
- SS23 Rulands: Non-equilibrium physics of machine learning (9 ECTS)
- SS23 Grün: Observing and Data Analysis Mthods for Cosmological Surveys (6 ECTS)
- WS23/24 Kepesidis: Deep Learning for Physicists
- WS23/24 Raeth: When machine learning meets complex systems
Seminars:
- SS23 Kepesidis: Artificial Intelligence in Scientific Research
- SS23 Kuhr: AI with and for physics
- WS23/24 Ensslin: Artificial Intelligence, Bayes, and Cognition
Lectures from Computer Science and Statistics:
- WS23/24 Bengs: Online Machine Learning and Bandits
- WS23/24 Bischl: Supervised Learning
- WS23/24 Bischl: Optimization
- WS23/24 Feurer: Machine Learning Operations
- WS23/24 Hüllermeier: Uncertainty in Artificial Intelligence and Machine Learning
- WS23/24 Linnhoff-Popien: Computational Intelligence
- WS23/24 Schubert: Deep Learning and Artificial Intelligence
- WS23/24 Seidl: Data Mining Algorithmen I
- WS23/24 Thomas: Automated Machine Learning
- WS23/24 Tresp: Machine Learning
- WS23/24 NN: Deep Learning for NLP
Study path (typical sequence)
- Semesters 1–2: Core physics + two ML lectures (CS/Math/Statistics)
- Semester 1–2: Physics-AI lectures and seminars (15 ECTS total)
- Semester 1-2: AI Lab (9 ECTS) with project-based work
- Semester 3-4: Practical Phase and Master’s thesis at the AI/physics interface
Tip: Plan your CS/Stats registrations early — some courses have application windows or moodle sign-ups.
Where this can take you
- Research & PhD: HEP/astrophysics, quantum information, condensed matter, biophysics, medical physics
- Industry & labs: Data science, AI engineering, computer vision, scientific software, HPC/analytics, health tech, climate/remote sensing
Contact
For questions about the specialization and course planning, please contact Dr. Michael Walther:
E-mail: Dr. Michael Walther.