Jes Frellsen — accepted AI-conference papers

2026

Position: Agentic AI systems should be making Bayes-consistent decisions

Theodore Papamarkou, Pierre Alquier, Matthias Bauer, Wray Buntine, Andrew Davison, Gintare Karolina Dziugaite +24

ICML 2026poster

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 consult, or how many resources to invest. While the usefulness and feasibility of Bayesian approaches remain unclear for L…

Cited by 0 Source ↗Scholar ↗

2026

Supervised Guidance Training for Infinite-Dimensional Diffusion Models

Elizabeth Baker, Alexander Denker, Jes Frellsen

ICML 2026poster

Score-based diffusion models have recently been extended to infinite-dimensional function spaces, with uses such as inverse problems arising from partial differential equations. In the Bayesian formulation of inverse problems, the aim is to sample from a posterior distribution over functions obtaine…

Cited by 0 Source ↗Scholar ↗

2025

Bayesian Circular Regression with von Mises Quasi-Processes

Yarden Cohen, Alexandre Khae Wu Navarro, Jes Frellsen, Richard E. Turner, Raziel Riemer, Ari Pakman

AISTATS 2025poster

The need for regression models to predict circular values arises in many scientific fields. In this work we explore a family of expressive and interpretable distributions over circle-valued random functions related to Gaussian processes targeting two Euclidean dimensions conditioned on the unit circ…

Cited by 0 Source ↗Scholar ↗

2025

Hyper-Transforming Latent Diffusion Models

Ignacio Peis, Batuhan Koyuncu, Isabel Valera, Jes Frellsen

ICML 2025poster

We introduce a novel generative framework for functions by integrating Implicit Neural Representations (INRs) and Transformer-based hypernetworks into latent variable models. Unlike prior approaches that rely on MLP-based hypernetworks with scalability limitations, our method employs a Transformer-b…

Cited by 0 Source ↗PDF ↗Scholar ↗

2025

Kinetic Langevin Diffusion for Crystalline Materials Generation

François R J Cornet, Federico Bergamin, Arghya Bhowmik, Juan Maria Garcia-Lastra, Jes Frellsen, Mikkel N. Schmidt

ICML 2025poster

Generative modeling of crystalline materials using diffusion models presents a series of challenges: the data distribution is characterized by inherent symmetries and involves multiple modalities, with some defined on specific manifolds. Notably, the treatment of fractional coordinates representing…

Cited by 1 Source ↗PDF ↗Scholar ↗

2025

Zero-shot protein stability prediction by inverse folding models: a free energy interpretation

Jes Frellsen, Maher M. Kassem, Tone Bengtsen, Lars Olsen, Kresten Lindorff-Larsen, Jesper Ferkinghoff-Borg +1

NeurIPS 2025poster

Inverse folding models have proven to be highly effective zero-shot predictors of protein stability. Despite this success, the link between the amino acid preferences of an inverse folding model and the free-energy considerations underlying thermodynamic stability remains incompletely understood. A…

Cited by 5 Source ↗code ↗Scholar ↗

2023

Adaptive Cholesky Gaussian Processes

Simon Bartels, Kristoffer Stensbo-Smidt, Pablo Moreno-Munoz, Wouter Boomsma, Jes Frellsen, Soren Hauberg

AISTATS 2023poster

We present a method to approximate Gaussian process regression models to large datasets by considering only a subset of the data. Our approach is novel in that the size of the subset is selected on the fly during exact inference with little computational overhead. From an empirical observation that…

Cited by 5 Source ↗PDF ↗code ↗Scholar ↗

2023

Explainability as statistical inference

Hugo Henri Joseph Senetaire, Damien Garreau, Jes Frellsen, Pierre-Alexandre Mattei

ICML 2023poster

A wide variety of model explanation approaches have been proposed in recent years, all guided by very different rationales and heuristics. In this paper, we take a new route and cast interpretability as a statistical inference problem. We propose a general deep probabilistic model designed to produc…

Cited by 4 Source ↗PDF ↗Scholar ↗

2023

Implicit Variational Inference for High-Dimensional Posteriors

Anshuk Uppal, Kristoffer Stensbo-Smidt, Wouter Boomsma, Jes Frellsen

NeurIPS 2023spotlight

In variational inference, the benefits of Bayesian models rely on accurately capturing the true posterior distribution. We propose using neural samplers that specify implicit distributions, which are well-suited for approximating complex multimodal and correlated posteriors in high-dimensional space…

Cited by 6 Source ↗PDF ↗code ↗Scholar ↗

2023

Learning To Generate 3d Representations of Building Roofs Using Single-View Aerial Imagery

Maxim Khomiakov, Alejandro Valverde Mahou, Alba Reinders Sánchez, Jes Frellsen, Michael Riis Andersen

ICASSP 2023accepted

We present a novel pipeline for learning the conditional distribution of a building roof mesh given pixels from an aerial image, under the assumption that roof geometry follows a set of regular patterns. Unlike alternative methods that require multiple images of the same object, our approach enables…

Cited by 0 Source ↗Scholar ↗

2023

That Label's got Style: Handling Label Style Bias for Uncertain Image Segmentation

Kilian Zepf, Eike Petersen, Jes Frellsen, Aasa Feragen

ICLR 2023poster

Segmentation uncertainty models predict a distribution over plausible segmentations for a given input, which they learn from the annotator variation in the training set. However, in practice these annotations can differ systematically in the way they are generated, for example through the use of dif…

Cited by 10 Source ↗PDF ↗Scholar ↗

2022

Exploring Predictive Uncertainty and Calibration in NLP: A Study on the Impact of Method & Data Scarcity

Dennis Ulmer, Jes Frellsen, Christian Hardmeier

EMNLP 2022finding

We investigate the problem of determining the predictive confidence (or, conversely, uncertainty) of a neural classifier through the lens of low-resource languages. By training models on sub-sampled datasets in three different languages, we assess the quality of estimates from a wide array of approa…

Cited by 23 Source ↗PDF ↗code ↗Scholar ↗

2022

How to deal with missing data in supervised deep learning?

Niels Bruun Ipsen, Pierre-Alexandre Mattei, Jes Frellsen

ICLR 2022poster

The issue of missing data in supervised learning has been largely overlooked, especially in the deep learning community. We investigate strategies to adapt neural architectures for handling missing values. Here, we focus on regression and classification problems where the features are assumed to be…

Cited by 50 Source ↗PDF ↗Scholar ↗

2022

Model-agnostic out-of-distribution detection using combined statistical tests

Federico Bergamin, Pierre-Alexandre Mattei, Jakob Drachmann Havtorn, Hugo Sénétaire, Hugo Schmutz, Lars Maaløe +2

AISTATS 2022poster

We present simple methods for out-of-distribution detection using a trained generative model. These techniques, based on classical statistical tests, are model-agnostic in the sense that they can be applied to any differentiable generative model. The idea is to combine a classical parametric test (R…

Cited by 26 Source ↗PDF ↗code ↗Scholar ↗

2021

Bounds all around: training energy-based models with bidirectional bounds

Cong Geng, Jia Wang, Zhiyong Gao, Jes Frellsen, Søren Hauberg

NeurIPS 2021poster

Energy-based models (EBMs) provide an elegant framework for density estimation, but they are notoriously difficult to train. Recent work has established links to generative adversarial networks, where the EBM is trained through a minimax game with a variational value function. We propose a bidirecti…

Cited by 19 Source ↗PDF ↗Scholar ↗

2021

Hierarchical VAEs Know What They Don’t Know

Jakob D. Havtorn, Jes Frellsen, Søren Hauberg, Lars Maaløe

ICML 2021spotlight

Deep generative models have been demonstrated as state-of-the-art density estimators. Yet, recent work has found that they often assign a higher likelihood to data from outside the training distribution. This seemingly paradoxical behavior has caused concerns over the quality of the attained density…

Cited by 87 Source ↗PDF ↗code ↗Scholar ↗

2021

not-MIWAE: Deep Generative Modelling with Missing not at Random Data

Niels Bruun Ipsen, Pierre-Alexandre Mattei, Jes Frellsen

ICLR 2021poster

When a missing process depends on the missing values themselves, it needs to be explicitly modelled and taken into account while doing likelihood-based inference. We present an approach for building and fitting deep latent variable models (DLVMs) in cases where the missing process is dependent on th…

Cited by 88 Source ↗PDF ↗code ↗Scholar ↗

2019

MIWAE: Deep Generative Modelling and Imputation of Incomplete Data Sets

Pierre-Alexandre Mattei, Jes Frellsen

ICML 2019oral

We consider the problem of handling missing data with deep latent variable models (DLVMs). First, we present a simple technique to train DLVMs when the training set contains missing-at-random data. Our approach, called MIWAE, is based on the importance-weighted autoencoder (IWAE), and maximises a po…

Cited by 356 Source ↗PDF ↗Scholar ↗

2019

Partially Exchangeable Networks and Architectures for Learning Summary Statistics in Approximate Bayesian Computation

Samuel Wiqvist, Pierre-Alexandre Mattei, Umberto Picchini, Jes Frellsen

ICML 2019oral

We present a novel family of deep neural architectures, named partially exchangeable networks (PENs) that leverage probabilistic symmetries. By design, PENs are invariant to block-switch transformations, which characterize the partial exchangeability properties of conditionally Markovian processes.…

Cited by 40 Source ↗PDF ↗code ↗Scholar ↗

2018

Leveraging the Exact Likelihood of Deep Latent Variable Models

Pierre-Alexandre Mattei, Jes Frellsen

NeurIPS 2018poster

Deep latent variable models (DLVMs) combine the approximation abilities of deep neural networks and the statistical foundations of generative models. Variational methods are commonly used for inference; however, the exact likelihood of these models has been largely overlooked. The purpose of this wo…

Cited by 77 Source ↗PDF ↗Scholar ↗

2017

Spherical convolutions and their application in molecular modelling

Wouter Boomsma, Jes Frellsen

NeurIPS 2017spotlight

Convolutional neural networks are increasingly used outside the domain of image analysis, in particular in various areas of the natural sciences concerned with spatial data. Such networks often work out-of-the box, and in some cases entire model architectures from image analysis can be carried over…

Cited by 112 Source ↗PDF ↗Scholar ↗

2016

Bayesian Generalised Ensemble Markov Chain Monte Carlo

Jes Frellsen, Ole Winther, Zoubin Ghahramani, Jesper Ferkinghoff-Borg

AISTATS 2016poster

Bayesian generalised ensemble (BayesGE) is a new method that addresses two major drawbacks of standard Markov chain Monte Carlo algorithms for inference in high-dimensional probability models: inapplicability to estimate the partition function and poor mixing properties. BayesGE uses a Bayesian appr…

Cited by 12 Source ↗PDF ↗Scholar ↗