Publications (29)
Understanding Probabilistic Sparse Gaussian Process Approximations
Matthias Bauer, Mark van der Wilk, Carl Edward Rasmussen
Good sparse approximations are essential for practical inference in Gaussian Processes as the computational cost of exact methods is prohibitive for large datasets. The Fully Indep…
Ecological feedback in quorum-sensing microbial populations can induce heterogeneous production of autoinducers
Matthias Bauer, Johannes Knebel, Matthias Lechner +2
Autoinducers are small signaling molecules that mediate intercellular communication in microbial populations and trigger coordinated gene expression via "quorum sensing". Elucidati…
Ultrafast two-colour X-ray emission spectroscopy reveals excited state landscape in a base metal dyad
Michal Nowakowski, Marina Huber-Gedert, Hossam Elgabarty +15
Effective photoinduced charge transfer makes molecular bimetallic assemblies attractive for applications as active light induced proton reduction systems. For a more sustainable fu…
Laplace Redux -- Effortless Bayesian Deep Learning
Erik Daxberger, Agustinus Kristiadi, Alexander Immer +3
Bayesian formulations of deep learning have been shown to have compelling theoretical properties and offer practical functional benefits, such as improved predictive uncertainty qu…
Generalized Doubly Reparameterized Gradient Estimators
Matthias Bauer, Andriy Mnih
Efficient low-variance gradient estimation enabled by the reparameterization trick (RT) has been essential to the success of variational autoencoders. Doubly-reparameterized gradie…
3-D Atomic Mapping of Interfacial Roughness and its Spatial Correlation Length in sub-10 nm Superlattices
Samik Mukherjee, Anis Attiaoui, Matthias Bauer +1
The interfacial abruptness and uniformity in heterostructures are critical to control their electronic and optical properties. With this perspective, this work demonstrates the 3-D…
Water structure near the surface of Weyl semimetals as catalysts in photocatalytic proton reduction
Jure Gujt, Peter Zimmer, Frederik Zysk +4
In this work, second-generation Car-Parrinello-based QM/MM molecular dynamics simulations of small nanoparticles of NbP, NbAs, TaAs and 1T-TaS in water are presented. The first…
Automatic Estimation of Modulation Transfer Functions
Matthias Bauer, Valentin Volchkov, Michael Hirsch +1
The modulation transfer function (MTF) is widely used to characterise the performance of optical systems. Measuring it is costly and it is thus rarely available for a given lens sp…
C3: High-performance and low-complexity neural compression from a single image or video
Hyunjik Kim, Matthias Bauer, Lucas Theis +2
Most neural compression models are trained on large datasets of images or videos in order to generalize to unseen data. Such generalization typically requires large and expressive…
Spatial Functa: Scaling Functa to ImageNet Classification and Generation
Matthias Bauer, Emilien Dupont, Andy Brock +3
Neural fields, also known as implicit neural representations, have emerged as a powerful means to represent complex signals of various modalities. Based on this Dupont et al. (2022…
Evaluating Numerical Reasoning in Text-to-Image Models
Ivana KajiÄ, Olivia Wiles, Isabela Albuquerque +4
Text-to-image generative models are capable of producing high-quality images that often faithfully depict concepts described using natural language. In this work, we comprehensivel…
Resampled Priors for Variational Autoencoders
Matthias Bauer, Andriy Mnih
We propose Learned Accept/Reject Sampling (LARS), a method for constructing richer priors using rejection sampling with a learned acceptance function. This work is motivated by rec…
Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
Gemini Team, Petko Georgiev, Ving Ian Lei +1132
In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over…
Finite-temperature Fe K-edge X-ray absorption simulations reveal local structural dynamics of an iron(II) photosensitizer in solution and the crystalline phase
Patrick Müller, Lorena Fritsch, Matthias Bauer +1
Interpreting metal K-edge spectra of flexible photosensitizers requires a structural model that separates electronic signatures from thermal motion, solvent disorder, and crystal-p…
Regularising for invariance to data augmentation improves supervised learning
Aleksander Botev, Matthias Bauer, Soham De
Data augmentation is used in machine learning to make the classifier invariant to label-preserving transformations. Usually this invariance is only encouraged implicitly by includi…
PaliGemma: A versatile 3B VLM for transfer
Lucas Beyer, Andreas Steiner, André Susano Pinto +32
PaliGemma is an open Vision-Language Model (VLM) that is based on the SigLIP-So400m vision encoder and the Gemma-2B language model. It is trained to be a versatile and broadly know…
Localized Energy States Induced by Atomic-Level Interfacial Broadening in Heterostructures
Anis Attiaoui, Gabriel Fettu, Samik Mukherjee +2
A theoretical framework incorporating atomic-level interfacial details is derived to include the electronic structure of buried interfaces and describe the behavior of charge carri…
Meta-Learning Probabilistic Inference For Prediction
Jonathan Gordon, John Bronskill, Matthias Bauer +2
This paper introduces a new framework for data efficient and versatile learning. Specifically: 1) We develop ML-PIP, a general framework for Meta-Learning approximate Probabilistic…
Emerging Markets for RFID Traces
Matthias Bauer, Benjamin Fabian, Matthias Fischmann +1
RFID tags are held to become ubiquitous in logistics in the near future, and item-level tagging will pave the way for Ubiquitous Computing, for example in application fields like s…
Discriminative k-shot learning using probabilistic models
Matthias Bauer, Mateo Rojas-Carulla, Jakub BartÅomiej ÅwiÄ tkowski +2
This paper introduces a probabilistic framework for k-shot image classification. The goal is to generalise from an initial large-scale classification task to a separate task compri…
Improving predictions of Bayesian neural nets via local linearization
Alexander Immer, Maciej Korzepa, Matthias Bauer
The generalized Gauss-Newton (GGN) approximation is often used to make practical Bayesian deep learning approaches scalable by replacing a second order derivative with a product of…
Interpretable and Differentially Private Predictions
Frederik Harder, Matthias Bauer, Mijung Park
Interpretable predictions, where it is clear why a machine learning model has made a particular decision, can compromise privacy by revealing the characteristics of individual data…
Scalable Marginal Likelihood Estimation for Model Selection in Deep Learning
Alexander Immer, Matthias Bauer, Vincent Fortuin +2
Marginal-likelihood based model-selection, even though promising, is rarely used in deep learning due to estimation difficulties. Instead, most approaches rely on validation data,…
New Covert Channels in HTTP
Matthias Bauer
This paper presents new methods enabling anonymous communication on the Internet. We describe a new protocol that allows us to create an anonymous overlay network by exploiting the…
Learning Invariances using the Marginal Likelihood
Mark van der Wilk, Matthias Bauer, ST John +1
Generalising well in supervised learning tasks relies on correctly extrapolating the training data to a large region of the input space. One way to achieve this is to constrain the…
Proofs of Zero Knowledge
Matthias Bauer
We present a protocol for verification of ``no such entry'' replies from databases. We introduce a new cryptographic primitive as the underlying structure, the keyed hash tree, whi…
Good, Cheap, and Fast: Overfitted Image Compression with Wasserstein Distortion
Jona Ballé, Luca Versari, Emilien Dupont +2
Inspired by the success of generative image models, recent work on learned image compression increasingly focuses on better probabilistic models of the natural image distribution,…
Position: agentic AI orchestration should be Bayes-consistent
Theodore Papamarkou, Pierre Alquier, Matthias Bauer +27
LLMs excel at predictive tasks and complex reasoning tasks, but many high-value deployments rely on decisions under uncertainty, for example, which tool to call, which expert to co…
Improving fine-grained understanding in image-text pre-training
Ioana Bica, Anastasija IliÄ, Matthias Bauer +8
We introduce SPARse Fine-grained Contrastive Alignment (SPARC), a simple method for pretraining more fine-grained multimodal representations from image-text pairs. Given that multi…