Rickmer Schulte @rickmer.bsky.social · 12/05/2026By examining what drives these recurring layer-wise ID patterns across neural network architectures, we offer a reinterpretation of this commonly observed phenomenon. 010
Rickmer Schulte @rickmer.bsky.social · 12/05/2026We revisit commonly reported layer-wise ID patterns and identify a notable mismatch between the theory and practice of ID estimation in this setting, showing that that common ID estimators do not track the true underlying ID of neural representations. 110
Rickmer Schulte @rickmer.bsky.social · 12/05/2026In this work, we take a closer look at intrinsic dimension (ID) estimation in neural representations. Motivated by ideas stemming from the manifold hypothesis, ID estimation has become a widely used tool for studying the internal representations of neural networks. 110
Rickmer Schulte @rickmer.bsky.social · 12/05/2026It was great to present and discuss our paper, “Rethinking Intrinsic Dimension Estimation in Neural Representations,” at #AISTATS 2026 last week🚀 📄 Paper: arxiv.org/abs/2604.20276 Joint work with @davidruegamer.bsky.social 1132
Rickmer Schulte @rickmer.bsky.social · 13/07/2025Special thanks to my co-authors @davidruegamer.bsky.social and @tnagler.bsky.social—really enjoyed this collaboration! 010
Rickmer Schulte @rickmer.bsky.social · 13/07/2025In this work, we explore how pre-trained neural networks can be leveraged to adjust for confounding in treatment effect estimation involving complex data such as images or text—providing a principled approach to integrate non-tabular data into causal effect estimation. 140
Rickmer Schulte @rickmer.bsky.social · 13/07/2025Looking forward to presenting our paper "Adjustment for Confounding using Pre-Trained Representations" at #ICML2025 in Vancouver next week! 🎉🇨🇦 Feel free to check out our paper and reach out if you're attending or would like to discuss! 📄 Paper: arxiv.org/abs/2506.14329 1101
Reposted by Rickmer SchulteMunich Center for Machine Learning @munichcenterml.bsky.social · 11/07/2025MCML researchers will be represented at #ICML2025 with more than 20 accepted paper 🎉 Check them out: mcml.ai/news/2025-07...mcml.aiMCML Researchers With 22 Papers at ICML 2025We are happy to announce that MCML researchers are represented with 22 papers at ICML 2025. 052
Rickmer Schulte @rickmer.bsky.social · 09/05/2025It has been a great honour to presented our paper "Additive Model Boosting: New Insights and Path(ologie)s" at #AISTATS 2025!🚀 Joint work with @davidruegamer.bsky.social! Check out the paper and feel free to reach out: 🔗 arxiv.org/abs/2503.05538 0101
Reposted by Rickmer SchulteDavid Rügamer @davidruegamer.bsky.social · 01/05/2025It seems that we have 3 accepted papers at ICML 2025 🔥 5142
Reposted by Rickmer SchulteTim G. J. Rudner @timrudner.bsky.social · 29/04/2025Congratulations to the #AABI2025 Workshop Track Outstanding Paper Award recipients! 0238
Reposted by Rickmer SchulteDavid Rügamer @davidruegamer.bsky.social · 23/04/2025Arriving in Singapore this afternoon 🛬 I'll attend #ICLR2025, #AABI2025, and #AISTATS2025 together with many of my students and collaborators to present our 2 orals, 5 posters, and 14 workshop contributions 🚀 Feel free to drop by! 1115
Reposted by Rickmer SchulteDavid Rügamer @davidruegamer.bsky.social · 22/01/2025🔥🔥 Two papers accepted at #AISTATS2025 💪 - Oral: *Additive Model Boosting: New Insights and Path(ologie)s* by @rickmer.bsky.social - Poster: *Paths and Ambient Spaces in Neural Loss Landscapes* from Daniel Dold, Julius Kobialka, Nicolai Palm, Emanuel Sommer, Oliver Dürr 0112