Munich Center for Machine Learning

YouTube video by MCML_Munich Center for Machine Learning Machine Learning for Climate Action - with researcher Kerstin Forster

Meet Kerstin Forster, researcher at #LMU & #MCML, using machine learning to cut emissions, boost renewable energy, and track corporate sustainability — driving strategies for a healthier planet.

youtu.be/9mYRnXsyvLg

Matthias Feurer, Thomas Bayes Fellow at #MCML & interim prof at #LMU, builds tools to make ML easier for scientists & more efficient for experts. Next semester he starts as junior prof at TU Dortmund—sharing in our interview how the fellowship helped him get there: mcml.ai/news/2025-09...

𝗠𝗖𝗠𝗟 𝗕𝗹𝗼𝗴: Ever rerun feature importance plots and get different “top features”? That randomness comes from how explanations sample data.

A #ICLR2025 paper by MCML researchers flips this: Compress Then Explain (CTE) makes model explanations more accurate.

📄 Read more: mcml.ai/news/2025-09...

#MICCAI2025 starts today and we are happy that #MCML researchers are represented with 24 papers! Congrats to our researchers! 🎉

Check them out: mcml.ai/news/2025-09...

YouTube video by MCML_Munich Center for Machine Learning Predicting Health with AI - with researcher Simon Schallmoser

Meet Simon Schallmoser, researcher at #LMU & #MCML, who uses machine learning to forecast conditions and personalize care — from detecting low blood sugar in diabetic drivers to paving the way for safer, proactive healthcare.

youtu.be/fZA5gJxX704

𝗠𝗖𝗠𝗟 𝗕𝗹𝗼𝗴: Identifying an object’s pose from partial data is like spotting a mug when only the handle is visible.
GCE-Pose, a new method by #MCML Members and collaborators, presented at CVPR 2025, takes on this challenge with finesse.

📄 More on our blog: mcml.ai/news/2025-09...

✈️🇺🇸 AI X-Change Spotlight: The #MCML 𝗔𝗜 𝗫-𝗖𝗵𝗮𝗻𝗴𝗲 𝗣𝗿𝗼𝗴𝗿𝗮𝗺 supports our Members in collaborating internationally with leading research institutions. PhD student Hannah Laus spent 11 weeks at Harvard University. She shares her experience in a new blogpost: mcml.ai/news/2025-09...

The event Comparing Notes on Trustworthy AI explores the rise of AI agents and personal assistants—and the ethical challenges they pose.

🎙 Gitta Kutyniok (LMU/MCML), Lena Kästner (Uni Bayreuth), Barbara Plank (LMU/MCML)

💡 September 25, 2025 · 5:30 pm

🔗 Register: www.eventbrite.de/e/comparing-...

We are hiring! Join the #MCML team as a Science Manager for acquisition of third-party funding. Application deadline: October 05, 2025.
Read more ➡️ portal.mytum.de/jobs/sonstig...

YouTube video by MCML_Munich Center for Machine Learning Robots Seeing in the Dark - with researcher Yannick Burkhardt

Meet Yannick Burkhardt, researcher at #TUM & #MCML, using event cameras that capture changes in light up to a million times per second – giving machines lightning-fast perception for robotics and self-driving cars.

youtu.be/QJD5qnIb7z4

YouTube video by MCML_Munich Center for Machine Learning 3D Machine Perception Beyond Vision - with researcher Riccardo Marin

Meet Riccardo Marin, researcher at #TUM & #MCML, exploring computer vision and 3D geometry. From detecting flaws in manufacturing to studying the human body, his work brings machine perception closer to human experience.

youtu.be/ngC3NHy6_4U

AI and Memory Workshop
On Oct 29 in Munich researchers from MCML BAdW and bidt meet to discuss how AI algorithms influence individual and collective memory with expert talks and networking #AI #Memory #MCML #BAdW #bidt

MCML and LAMARR join forces in Bonn Sept 24–25 to explore generative AI and NLP with talks from PhD students postdocs and senior researchers plus networking to build collaborations #AI #NLP #MCML #LAMARR

Meet Clara Sophie Vetter, PhD student at #LMU & #MCML, who uses AI to find hidden patterns in brain scans, genetic data & smartphone data 🧠. Her goal: identify biomarkers to predict mental health risks & enable more personalized psychiatric care.

youtu.be/MPjdJ11eL6k

YouTube video by MCML_Munich Center for Machine Learning Satellite Insights for a Sustainable Future - with researcher Ivica Obadic

Meet Ivica Obadic, PhD student at #TUM & #MCML, who develops transparent AI models to analyze satellite images 🌍. His work detects and predicts urban changes, giving city planners explainable recommendations to design more sustainable, livable cities.

youtu.be/Qlur0GgsXcA

YouTube video by MCML_Munich Center for Machine Learning Digital Twins for Surgery - with researcher Azade Farshad

Meet Azade Farshad who develops digital twins of patients at #TUM & #MCML to improve treatment, surgical planning & training. Using graph-based analysis and multimodal data, she creates realistic surgical simulations to preview procedures & optimize strategies.

www.youtube.com/watch?v=9D1Q...

YouTube video by MCML_Munich Center for Machine Learning From Physics Dreams to Algorithm Discovery - with Prof. Niki Kilbertus

🎥 As a child, Niki Kilbertus wanted to discover a law of nature. Today, as a #TUM professor and PI at #Helmholtz & #MCML, he develops AI that doesn’t just spot patterns — it reveals cause and effect.

www.youtube.com/watch?v=EYut...

𝗠𝗖𝗠𝗟 𝗕𝗹𝗼𝗴: Tracking actions, not just objects — the Spatiotemporal Action Grounding Challenge & Workshop at ICCV 2025 tackles detecting when and where actions occur in complex videos.

📄 Read more: mcml.ai/news/2025-08...

YouTube video by MCML_Munich Center for Machine Learning AI for Dynamic Urban Mapping - with researcher Shanshan Bai

Imagine a “living map” of cities built from geo-tagged social media data. Meet Shanshan Bai ( #MCML/ #TUM) using #AI to identify building functions from online context—offering real-time insights for urban planning & emergency response when official data is outdated

www.youtube.com/watch?v=oAet...

𝗠𝗖𝗠𝗟 𝗕𝗹𝗼𝗴: Fine-tuning text-to-image models just got easier. A method by MCML members Vincent Tao Hu and Björn Ommer as well as team enables smooth control over attributes like age—without changing the rest of the image.
Read more in our blog🔗 mcml.ai/news/2025-08...

YouTube video by MCML_Munich Center for Machine Learning What is intelligence—and what kind of intelligence do we want in our future? With Prof. Sven Nyholm

🎥 Who gets the credit, or the blame, when AI makes decisions? Sven Nyholm ( #LMU/ #MCML) reflects on how AI challenges our ideas of agency, credit, and blame — and why we need new ways of thinking about authorship, justice, and decision-making.

www.youtube.com/watch?v=KUqi...

#AI #Ethics

YouTube video by MCML_Munich Center for Machine Learning AI for better Social Media - with researcher Dominik Bär

Meet Dominik Bär, #MCML junior member & PhD student at #LMU. He’s building AI that not only detects hate & misinformation but also understands why it's posted—and replies with thoughtful, real-time counterspeech.

www.youtube.com/watch?v=oTkB...

𝗠𝗖𝗠𝗟 𝗕𝗹𝗼𝗴: #MCML members Lukas Gosch, Stephan Günnemann, Debarghya Ghoshdastidar, as welll as Mahalakshmi Sabanayagam show exact guarantees on neural network stability under label flips.

📖 Read more: mcml.ai/news/2025-07...

YouTube video by MCML_Munich Center for Machine Learning Tracking Our Changing Planet from Space - with Prof. Xiaoxiang Zhu

🎥 From wanting to see Earth from space to building AI systems that help us understand it better. Xiaoxiang Zhu (TUM/#MCML) uses satellite imagery and Data Science to extract reliable geo-information.

www.youtube.com/watch?v=acO8...

#AI #EarthObservation

YouTube video by MCML_Munich Center for Machine Learning AI for Enhanced Eye Diagnostics - with researcher Lucie Huang

💡 How is #AI changing eye care? Lucie Huang, #MCML junior member + PhD student at #TUM, is developing AI for faster, sharper eye scans — enabling earlier diagnosis + better treatment.

🎥 First video in our KI Trans series w/ TüftelLab + LMU!
www.youtube.com/watch?v=mSm1...

MCML researchers with more than 20 papers at #ACL2025 in Vienna🎉

Check them out: mcml.ai/news/2025-07...

Recently, we visited SAP Labs in Garching and had an event of our series “MCML Pitchtalks with Industry”

Read more: mcml.ai/news/2025-07...

Thanks a lot for hosting us!

Conversations and new connections at our #MCML Stammtisch, yesterday🍕🍺

𝗠𝗖𝗠𝗟 𝗕𝗹𝗼𝗴: What if a single image could reveal an entire 3D scene—even the hidden parts? Humans do this instinctively. Inspired by this, #MCML Members and collaborators, developed SceneDINO🦖: a feed-forward method for unsupervised semantic scene completion.

📄 Read more: mcml.ai/news/2025-07...

YouTube video by MCML_Munich Center for Machine Learning How Reliable Are Machine Learning Methods? - Prof. Anne-Laure Boulesteix & PhD-Student Milena Wünsch

🎥 New algorithms often claim to be faster or more accurate — but hidden biases can distort results. At #LMU and #MCML, Anne-Laure Boulesteix and Milena Wünsch investigate how small choices in method comparisons can lead to big distortions — often without us noticing.

www.youtube.com/watch?v=x0jJ...