Pinned post
This is the #MachineLearning Group of Prof. Klaus-Robert Müller at the @TUBerlin. Let’s connect! #ScienceTwitter
ML Group, TU Berlin
@ml_tuberlin
This is the Machine Learning Group of Prof. Klaus-Robert Müller @TUBerlin. Proud part of @bifoldberlin.
70 Following 632 Followers 72 Posts
Latest posts 5
🔊
We have a new preprint on the consistency of representational similarities in #ComputerVision models like #ImageNet !
Check it out! 🔬
#MachineLearning #AI
We have a new preprint on the consistency of representational similarities in #ComputerVision models like #ImageNet !
Check it out! 🔬
#MachineLearning #AI
If two models are more similar to each other than a third on ImageNet, will this hold for medical/ satellite images? Our preprint analyzes how vision model similarities generalize across datasets, the factors that influence them, and their link to downstream task behavior. 🧵1/7View quoted post on X
🔊
Check out one of our latest #NeurIPS2024 publications on #XAI for state space sequence models!
#MachineLearning #AI
Check out one of our latest #NeurIPS2024 publications on #XAI for state space sequence models!
#MachineLearning #AI
🎉 Thrilled to announce that our paper MambaLRP has been accepted at #NeurIPS2024!
Explain your Mamba models in a faithful, efficient, and effective way! 🐍
Huge thanks to @EberleOliver, Grégoire Montavon, and Klaus-Robert Müller @bifoldberlin, @ml_tuberlin , @GoogleDeepMind,View quoted post on X
🔊
Check out our latest publication „Historical insights at scale: A corpus-wide machine learning analysis of early modern astronomic tables“. Now out in @ScienceAdvances 🎉
Together with @bifoldberlin @MPIWG
#MachineLearning #AI #History
Check out our latest publication „Historical insights at scale: A corpus-wide machine learning analysis of early modern astronomic tables“. Now out in @ScienceAdvances 🎉
Together with @bifoldberlin @MPIWG
#MachineLearning #AI #History
Our work on historical insights at scale using machine learning is now out in @ScienceAdvances! Very proud of this team effort, bridging disciplines and institutions—@MPIWG @TUBerlin @bifoldberlin @ml_tuberlin
📜science.org/doi/10.1126/sc…View quoted post on X
🔊
Finally online and accepted at #NeurIPS24!
Congratulations to the whole team 🎉
#MachineLearning #AI #ComputerVision #HumanAlignment
Finally online and accepted at #NeurIPS24!
Congratulations to the whole team 🎉
#MachineLearning #AI #ComputerVision #HumanAlignment
Which tasks benefit the most from aligning vision models with human perceptual judgments? In our most recent NeurIPS paper we pin down the downstream tasks where perceptual alignment yields the strongest increases in performance. More in the thread below 👇View quoted post on X
