97 citations · 164 across the 14 of their papers we have counts for
15 papers
Economical Jet Taggers -- Equivariant, Slim, and Quantized
Antoine Petitjean, Tilman Plehn, Jonas Spinner +1
Modern machine learning is transforming jet tagging at the LHC, but the leading transformer architectures are large, not particularly fast, and training-intensive. We present a sli…
OOD Detection with immature Models
Behrooz Montazeran, Ullrich Köthe
Likelihood-based deep generative models (DGMs) have gained significant attention for their ability to approximate the distributions of high-dimensional data. However, these models…
Analyzing Generative Models by Manifold Entropic Metrics
Daniel Galperin, Ullrich Köthe
Good generative models should not only synthesize high quality data, but also utilize interpretable representations that aid human understanding of their behavior. However, it is d…
Detecting Model Misspecification in Amortized Bayesian Inference with Neural Networks: An Extended Investigation
Marvin Schmitt, Paul-Christian Bürkner, Ullrich Köthe +1
Recent advances in probabilistic deep learning enable efficient amortized Bayesian inference in settings where the likelihood function is only implicitly defined by a simulation pr…
Towards Context-Aware Domain Generalization: Understanding the Benefits and Limits of Marginal Transfer Learning
Jens Müller, Lars Kühmichel, Martin Rohbeck +2
In this work, we analyze the conditions under which information about the context of an input can improve the predictions of deep learning models in new domains. Following work…
Consistency Models for Scalable and Fast Simulation-Based Inference
Marvin Schmitt, Valentin Pratz, Ullrich Köthe +2
Simulation-based inference (SBI) is constantly in search of more expressive and efficient algorithms to accurately infer the parameters of complex simulation models. In line with t…
Leveraging Self-Consistency for Data-Efficient Amortized Bayesian Inference
Marvin Schmitt, Desi R. Ivanova, Daniel Habermann +3
We propose a method to improve the efficiency and accuracy of amortized Bayesian inference by leveraging universal symmetries in the joint probabilistic model of parameters and dat…
BayesFlow: Amortized Bayesian Workflows With Neural Networks
Stefan T Radev, Marvin Schmitt, Lukas Schumacher +5
Modern Bayesian inference involves a mixture of computational techniques for estimating, validating, and drawing conclusions from probabilistic models as part of principled workflo…
JANA: Jointly Amortized Neural Approximation of Complex Bayesian Models
Stefan T. Radev, Marvin Schmitt, Valentin Pratz +3
This work proposes ``jointly amortized neural approximation'' (JANA) of intractable likelihood functions and posterior densities arising in Bayesian surrogate modeling and simulati…
Neural Superstatistics for Bayesian Estimation of Dynamic Cognitive Models
Lukas Schumacher, Paul-Christian Bürkner, Andreas Voss +2
Mathematical models of cognition are often memoryless and ignore potential fluctuations of their parameters. However, human cognition is inherently dynamic. Thus, we propose to aug…
Towards Multimodal Depth Estimation from Light Fields
Titus Leistner, Radek Mackowiak, Lynton Ardizzone +2
Light field applications, especially light field rendering and depth estimation, developed rapidly in recent years. While state-of-the-art light field rendering methods handle semi…
Detecting Model Misspecification in Amortized Bayesian Inference with Neural Networks
Marvin Schmitt, Paul-Christian Bürkner, Ullrich Köthe +1
Neural density estimators have proven remarkably powerful in performing efficient simulation-based Bayesian inference in various research domains. In particular, the BayesFlow fram…
Measuring QCD Splittings with Invertible Networks
Sebastian Bieringer, Anja Butter, Theo Heimel +4
QCD splittings are among the most fundamental theory concepts at the LHC. We show how they can be studied systematically with the help of invertible neural networks. These networks…
Generative Classifiers as a Basis for Trustworthy Image Classification
Radek Mackowiak, Lynton Ardizzone, Ullrich Köthe +1
With the maturing of deep learning systems, trustworthiness is becoming increasingly important for model assessment. We understand trustworthiness as the combination of explainabil…
Invertible Networks or Partons to Detector and Back Again
Marco Bellagente, Anja Butter, Gregor Kasieczka +5
For simulations where the forward and the inverse directions have a physics meaning, invertible neural networks are especially useful. A conditional INN can invert a detector simul…