Mattes Mollenhauer

New preprint! "Regularized least squares learning with heavy-tailed noise is minimax optimal" joint work with @moskitos_bite, @DimitriMeunier1 and @ArthurGretton

Wrapping up ICML - had a blast, great to see so many people again! @MartinGenzel @patrickputzky presenting our work on variable size model compression at ES-FoMO workshop😎

📢 Now it’s out on arXiv! If you’re interested in theoretical guarantees for learning with dependent Hilbert-space data, check out our full paper: 📝 An Empirical Bernstein Inequality for Dependent Data in Hilbert Spaces and Applications 👉 arxiv.org/abs/2507.07826

Come see how to shrink protein sequences with diffusion at our talk tomorrow morning!!!

I’m in Vancouver! Reach out if you want to grab a coffee and talk learning theory, model compression, Markov processes, inverse problems… We’ll also present a poster at Efficient Systems for Foundation Models Workshop!

You don’t need kindergarten to be a great AI researcher, but I would recommend preschool.

👏 Big shout out to all co-authors for an amazing collab: @patrickputzky, Pengfei Zhao, Sebastian Schulze, @gaussianmeasure, Robert Seidel, Stefan Dietzel, and Thomas Wollmann 📄 Paper: arxiv.org/abs/2502.01717 🧑‍💻 Code: github.com/merantix-momen… 🤗 Models: tinyurl.com/bdacy658

New preprint + code! Neural network compression to any compression rate without re-computation!

We've written a monograph on Gaussian processes and reproducing kernel methods. arxiv.org/abs/2506.17366

NeurIPS D&B track in a nutshell: (1) An LLM-generated benchmark dataset (2) used to test performance of LLMs (3) evaluated via LLM-as-a-judge

to provide some substance to the whole reasoning debate

Characterizing finely the decay of eigenvalues of kernel matrices: many people need it, but explicit references are hard to find. This blog post reviews amazing asymptotic results from Harold Widom (1963!) and proposes new non-asymptotic bounds. francisbach.com/spectrum-kerne…

Optimality and Adaptivity of Deep Neural Features for Instrumental Variable Regression ift.tt/8m1M4pd

Optimal Convergence Rates for Neural Operators ift.tt/JVzpStl

Truly excellent piece on entropy. Source: Quanta Magazine search.app/ichMj5mCxENsWC…

Happy to see such issues raised and discussed in a NeurIPS workshop. I highly recommend checking the slides at: cims.nyu.edu/~matus/neurips…

In his book “The Nature of Statistical Learning” V. Vapnik wrote: “When solving a given problem, try to avoid a more general problem as an intermediate step”