AMLab | Amsterdam Machine Learning Lab

Jan-Willem van de Meent

Associate professor (UHD)
AMLab and Delta Lab
Informatics Institute
University of Amsterdam
Science Park, Lab 42, L4.13

Google scholar Github Mastodon Twitter

Jan-Willem van de Meent is an Associate Professor (Universitair Hoofddocent) at the University of Amsterdam, where he directs the AMLab, co-directs the UvA Bosch Delta Lab, and directs the Amsterdam ELLIS Unit.

His research develops principled methods for generative AI and probabilistic inference, with a focus on scalable and data-efficient scientific computation. This includes scalable neural architectures for physical systems, learned surrogate models, and deep generative models for molecular and materials design. An ongoing interest is variational flow matching, a probabilistic perspective on flow-based generative models that informs principled approaches to discrete and structured data, controlled generation, and test-time conditioning. He collaborates closely with experts in physical chemistry, fluid mechanics, and materials science.

Earlier in his career he worked extensively on probabilistic programming, the design of languages and inference algorithms that allow researchers to express complex probabilistic models as programs. This led to the development of Anglican, foundational work on inference methods for universal probabilistic programs, and a textbook.

Prior to joining the University of Amsterdam, he was an Assistant Professor at Northeastern University. He carried out his PhD research in biophysics at Leiden and Cambridge with Wim van Saarloos and Ray Goldstein, and held postdoctoral positions with Frank Wood at Oxford and with Chris Wiggins and Ruben Gonzalez at Columbia. He was a founding co-chair of the international conference on probabilistic programming (PROBPROG) and a program chair for AISTATS 2023.

Recent Publications

2026

arXiv

Categorical Flow Maps

Roos, Daan, Davis, Oscar, Eijkelboom, Floor, Bronstein, Michael, Welling, Max, Ceylan, İsmail İlkan, Ambrogioni, Luca, and van de Meent, Jan-Willem

arXiv preprint 2026

HTML PDF
ICLR

Purrception: Variational Flow Matching for Vector-Quantized Image Generation

Matişan, Răzvan-Andrei, Hu, Vincent Tao, Bartosh, Grigory, Ommer, Björn, Snoek, Cees G. M., Welling, Max, van de Meent, Jan-Willem, Derakhshani, Mohammad Mahdi, and Eijkelboom, Floor

In International Conference on Learning Representations 2026

HTML PDF
ICLR

CORDS: Continuous Representations of Discrete Structures

Hadži Veljković, Tin, Bekkers, Erik J., Tiemann, Michael, and van de Meent, Jan-Willem

In International Conference on Learning Representations 2026

HTML PDF
ICLR

Riemannian Variational Flow Matching for Material and Protein Design

Zaghen, Olga, Eijkelboom, Floor, Pouplin, Alison, Liu, Cong, Welling, Max, van de Meent, Jan-Willem, and Bekkers, Erik J.

In International Conference on Learning Representations 2026

HTML PDF

2025

arXiv

MSPT: Efficient Large-Scale Physical Modeling via Parallelized Multi-Scale Attention

Curvo, Pedro M. P., van de Meent, Jan-Willem, and Zhdanov, Maksim

arXiv preprint 2025

HTML PDF
arXiv

Discovering Lie Groups with Flow Matching

Park, Jung Yeon, Chen, Yuxuan, Eijkelboom, Floor, van de Meent, Jan-Willem, Wong, Lawson L. S., and Walters, Robin

arXiv preprint 2025

HTML PDF
ICML

Controlled Generation with Equivariant Variational Flow Matching

Eijkelboom, Floor, Zimmermann, Heiko, Bekkers, Erik, Welling, Max, Naesseth, Christian, and van de Meent, Jan-Willem

In International Conference on Machine Learning 2025

HTML PDF
ICML

Exponential Family Variational Flow Matching for Tabular Data Generation

Guzmán-Cordero*, Andrés, Eijkelboom*, Floor, and van de Meent, Jan-Willem

In International Conference on Machine Learning 2025

HTML PDF
ICML

Erwin: A Tree-based Hierarchical Transformer for Large-scale Physical Systems

Zhdanov, Maksim, Welling, Max, and van de Meent, Jan-Willem

In International Conference on Machine Learning 2025

Abs HTML PDF Code

Large-scale physical systems defined on irregular grids pose significant scalability challenges for deep learning methods, especially in the presence of long-range interactions and multi-scale coupling. Traditional approaches that compute all pairwise interactions, such as attention, become computationally prohibitive as they scale quadratically with the number of nodes. We present Erwin, a hierarchical transformer inspired by methods from computational many-body physics, which combines the efficiency of tree-based algorithms with the expressivity of attention mechanisms. Erwin employs ball tree partitioning to organize computation, which enables linear-time attention by processing nodes in parallel within local neighborhoods of fixed size. Through progressive coarsening and refinement of the ball tree structure, complemented by a novel cross-ball interaction mechanism, it captures both fine-grained local details and global features. We demonstrate Erwin’s effectiveness across multiple domains, including cosmology, molecular dynamics, and particle fluid dynamics, where it consistently outperforms baseline methods both in accuracy and computational efficiency.
PRL

Learning Neural Free-Energy Functionals with Pair-Correlation Matching

Dijkman, Jacobus, Dijkstra, Marjolein, van Roij, René, Welling, Max, van de Meent, Jan-Willem, and Ensing, Bernd

Physical Review Letters Feb 2025

Abs HTML PDF

The intrinsic Helmholtz free-energy functional, the centerpiece of classical density functional theory, is at best only known approximately for 3D systems. Here we introduce a method for learning a neural-network approximation of this functional by exclusively training on a dataset of radial distribution functions, circumventing the need to sample costly heterogeneous density profiles in a wide variety of external potentials. For a supercritical Lennard-Jones system with planar symmetry, we demonstrate that the learned neural free-energy functional accurately predicts inhomogeneous density profiles under various complex external potentials obtained from simulations.
ICRA

On-Robot Reinforcement Learning with Goal-Contrastive Rewards

Biza, Ondrej, Weng, Thomas, Sun, Lingfeng, Schmeckpeper, Karl, Kelestemur, Tarik, Ma, Yecheng Jason, Platt, Robert, van de Meent, Jan-Willem, and Wong, Lawson L. S.

In Proceedings of the 2025 IEEE International Conference on Robotics and Automation, ICRA’25 Feb 2025

HTML PDF

2024

NeurIPS

Practical Shuffle Coding

Kunze, Julius, Severo, Daniel, van de Meent, Jan-Willem, and Townsend, James

In Advances in Neural Information Processing Systems 2024

HTML PDF Code
NeurIPS

Variational Flow Matching for Graph Generation

Eijkelboom, Floor, Bartosh, Grigory, Naesseth, Christian A., Welling, Max, and van de Meent, Jan-Willem

In Advances in Neural Information Processing Systems 2024

HTML PDF
NeurIPS

VISA: Variational Inference with Sequential Sample-Average Approximations

Zimmermann, Heiko, Naesseth, Christian A., and van de Meent, Jan-Willem

In Advances in Neural Information Processing Systems 2024

HTML PDF Code
NAACL

Towards Reducing Diagnostic Errors with Interpretable Risk Prediction

McInerney, Denis Jered, Dickinson, William, Flynn, Lucy C., Young, Andrea C., Young, Geoffrey S., van de Meent, Jan-Willem, and Wallace, Byron C.

In 2024 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL) 2024

Abs HTML PDF

Many diagnostic errors occur because clinicians cannot easily access relevant information in patient Electronic Health Records (EHRs). In this work we propose a method to use LLMs to identify pieces of evidence in patient EHR data that indicate increased or decreased risk of specific diagnoses; our ultimate aim is to increase access to evidence and reduce diagnostic errors. In particular, we propose a Neural Additive Model to make predictions backed by evidence with individualized risk estimates at time-points where clinicians are still uncertain, aiming to specifically mitigate delays in diagnosis and errors stemming from an incomplete differential. To train such a model, it is necessary to infer temporally fine-grained retrospective labels of eventual "true" diagnoses. We do so with LLMs, to ensure that the input text is from before a confident diagnosis can be made. We use an LLM to retrieve an initial pool of evidence, but then refine this set of evidence according to correlations learned by the model. We conduct an in-depth evaluation of the usefulness of our approach by simulating how it might be used by a clinician to decide between a pre-defined list of differential diagnoses.
ICLR

Entropy Coding of Unordered Data Structures

Kunze, Julius, Severo, Daniel, Zani, Giulio, van de Meent, Jan-Willem, and Townsend, James

In International Conference on Learning Representations (ICLR) 2024

HTML PDF Code

2023

EMNLP

CHiLL: Zero-shot Custom Interpretable Feature Extraction from Clinical Notes with Large Language Models

McInerney, Denis Jered, Young, Geoffrey, Meent, Jan-Willem, and Wallace, Byron

In The 2023 Conference on Empirical Methods in Natural Language Processing 2023

HTML PDF
NeurIPS

Topological Obstructions and How to Avoid Them

Esmaeili, Babak, Walters, Robin, Zimmermann, Heiko, and van de Meent, Jan-Willem

In Thirty-seventh Conference on Neural Information Processing Systems (to appear) Dec 2023

HTML PDF
CoRL

One-shot Imitation Learning via Interaction Warping

Biza, Ondrej, Thompson, Skye, Pagidi, Kishore Reddy, Kumar, Abhinav, Pol, Elise, Walters, Robin, Kipf, Thomas, Meent, Jan-Willem, Wong, Lawson L.S., and Platt, Robert

In 7th Annual Conference on Robot Learning Nov 2023

HTML PDF
ACT

String Diagrams with Factorized Densities

Sennesh, Eli, and van de Meent, Jan-Willem

In Applied Category Theory Jul 2023

Abs HTML PDF

A growing body of research on probabilistic programs and causal models has highlighted the need to reason compositionally about model classes that extend directed graphical models. Both probabilistic programs and causal models define a joint probability density over a set of random variables, and exhibit sparse structure that can be used to reason about causation and conditional independence. This work builds on recent work on Markov categories of probabilistic mappings to define a category whose morphisms combine a joint density, factorized over each sample space, with a deterministic mapping from samples to return values. This is a step towards closing the gap between recent category-theoretic descriptions of probability measures, and the operational definitions of factorized densities that are commonly employed in probabilistic programming and causal inference.
TMLR

A Variational Perspective on Generative Flow Networks

Zimmermann, Heiko, Lindsten, Fredrik, Meent, Jan-Willem, and Naesseth, Christian A

Transactions on Machine Learning Research Apr 2023

Abs HTML PDF Code

Generative flow networks (GFNs) are a class of probabilistic models for sequential sampling of composite objects, proportional to a target distribution that is defined in terms of an energy function or a reward. GFNs are typically trained using a flow matching or trajectory balance objective, which matches forward and backward transition models over trajectories. In this work we introduce a variational objective for training GFNs, which is a convex combination of the reverse- and forward KL divergences, and compare it to the trajectory balance objective when sampling from the forward- and backward model, respectively. We show that, in certain settings, variational inference for GFNs is equivalent to minimizing the trajectory balance objective, in the sense that both methods compute the same score-function gradient. This insight suggests that in these settings, control variates, which are commonly used to reduce the variance of score-function gradient estimates, can also be used with the trajectory balance objective. We evaluate our findings and the performance of the proposed variational objective numerically by comparing it to the trajectory balance objective on two synthetic tasks.
LAFI

Verified Reversible Programming for Verified Lossless Compression

Townsend, James, and van de Meent, Jan-Willem

In POPL Workshop on Languages for Inference (LAFI) Apr 2023

HTML PDF
LAFI

A convenient category of tracing measure kernels

Sennesh, Eli, and van de Meent, Jan-Willem

In POPL Workshop on Languages for Inference (LAFI) Apr 2023

HTML PDF
LAFI

Semantics of Probabilistic Program Traces

Lew, Alexander, Sennesh, Eli, van de Meent, Jan-Willem, and Mansinghka, Vikash

In POPL Workshop on Languages for Inference (LAFI) Apr 2023

HTML PDF

2022

EMNLP

That’s the Wrong Lung! Evaluating and Improving the Interpretability of Unsupervised Multimodal Encoders for Medical Data

McInerney, Denis Jered, Young, Geoffrey, van de Meent, Jan-Willem, and Wallace, Byron C.

Proceedings of the Conference on Empirical Methods in Natural Language Processing. Conference on Empirical Methods in Natural Language Processing Dec 2022

Abs

Pretraining multimodal models on Electronic Health Records (EHRs) provides a means of learning representations that can transfer to downstream tasks with minimal supervision. Recent multimodal models induce soft local alignments between image regions and sentences. This is of particular interest in the medical domain, where alignments might highlight regions in an image relevant to specific phenomena described in free-text. While past work has suggested that attention “heatmaps” can be interpreted in this manner, there has been little evaluation of such alignments. We compare alignments from a state-of-the-art multimodal (image and text) model for EHR with human annotations that link image regions to sentences. Our main finding is that the text has an often weak or unintuitive influence on attention; alignments do not consistently reflect basic anatomical information. Moreover, synthetic modifications — such as substituting “left” for “right” — do not substantially influence highlights. Simple techniques such as allowing the model to opt out of attending to the image and few-shot finetuning show promise in terms of their ability to improve alignments with very little or no supervision. We make our code and checkpoints open-source.
NeurIPS WSGNR

Understanding Optimization Challenges when Encoding to Geometric Structures

Esmaeili, Babak, Walters, Robin, Zimmermann, Heiko, and Meent, Jan-Willem

In NeurIPS 2022 Workshop on Symmetry and Geometry in Neural Representations Dec 2022
PLOS Comp Bio

Probabilistic Program Inference in Network-Based Epidemiological Simulations

Smedemark-Margulies, Niklas, Walters, Robin, Zimmermann, Heiko, Laird, Lucas, Loo, Christian, Kaushik, Neela, Caceres, Rajmonda, and Meent, Jan-Willem

PLOS Computational Biology Nov 2022

Abs HTML PDF

Accurate epidemiological models require parameter estimates that account for mobility patterns and social network structure. We demonstrate the effectiveness of probabilistic programming for parameter inference in these models. We consider an agent-based simulation that represents mobility networks as degree-corrected stochastic block models, whose parameters we estimate from cell phone co-location data. We then use probabilistic program inference methods to approximate the distribution over disease transmission parameters conditioned on reported cases and deaths. Our experiments demonstrate that the resulting models improve the quality of fit in multiple geographies relative to baselines that do not model network topology.
Bio. Pysch.

Interoception as Modeling, Allostasis as Control

Sennesh, Eli, Theriault, Jordan, Brooks, Dana, Meent, Jan-Willem, Barrett, Lisa Feldman, and Quigley, Karen S.

Biological Psychology Nov 2022

Abs HTML PDF

The brain regulates the body by anticipating its needs and attempting to meet them before they arise – a process called allostasis. Allostasis requires a model of the changing sensory conditions within the body, a process called interoception. In this paper, we examine how interoception may provide performance feedback for allostasis. We suggest studying allostasis in terms of control theory, reviewing control theory’s applications to related issues in physiology, motor control, and decision making. We synthesize these by relating them to the important properties of allostatic regulation as a control problem. We then sketch a novel formalism for how the brain might perform allostatic control of the viscera by analogy to skeletomotor control, including a mathematical view on how interoception acts as performance feedback for allostasis. Finally, we suggest ways to test implications of our hypotheses.
Neuroinf.

A Computational Neural Model for Mapping Degenerate Neural Architectures

Khan, Zulqarnain, Wang, Yiyu, Sennesh, Eli, Dy, Jennifer, Ostadabbas, Sarah, van de Meent, Jan-Willem, Hutchinson, J. Benjamin, and Satpute, Ajay B.

Neuroinformatics Mar 2022

Abs HTML PDF

Degeneracy in biological systems refers to a many-to-one mapping between physical structures and their functional (including psychological) outcomes. Despite the ubiquity of the phenomenon, traditional analytical tools for modeling degeneracy in neuroscience are extremely limited. In this study, we generated synthetic datasets to describe three situations of degeneracy in fMRI data to demonstrate the limitations of the current univariate approach. We describe a novel computational approach for the analysis referred to as neural topographic factor analysis (NTFA). NTFA is designed to capture variations in neural activity across task conditions and participants. The advantage of this discovery-oriented approach is to reveal whether and how experimental trials and participants cluster into task conditions and participant groups. We applied NTFA on simulated data, revealing the appropriate degeneracy assumption in all three situations and demonstrating NTFA’s utility in uncovering degeneracy. Lastly, we discussed the importance of testing degeneracy in fMRI data and the implications of applying NTFA to do so.
WACV

Enhancing Few-Shot Image Classification With Unlabelled Examples

Bateni, Peyman, Barber, Jarred, Meent, Jan-Willem, and Wood, Frank

In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Jan 2022

Abs HTML PDF Code

We develop a transductive meta-learning method that uses unlabelled instances to improve few-shot image classification performance. Our approach combines a regularized Mahalanobis-distance-based soft k-means clustering procedure with a modified state of the art neural adaptive feature extractor to achieve improved test-time classification accuracy using unlabelled data. We evaluate our method on transductive few-shot learning tasks, in which the goal is to jointly predict labels for query (test) examples given a set of support (training) examples. We achieve state of the art performance on the Meta-Dataset, mini-ImageNet and tiered-ImageNet benchmarks. All trained models and code have been made publicly available

2021

AISTATS

Rate-Regularization and Generalization in Variational Autoencoders

Bozkurt, Alican, Esmaeili, Babak, Tristan, Jean-Baptiste, Brooks, Dana, Dy, Jennifer, and Meent, Jan-Willem

In International Conference on Artificial Intelligence and Statistics Mar 2021

Abs HTML PDF

Variational autoencoders (VAEs) optimize an objective that comprises a reconstruction loss (the distortion) and a KL term (the rate). The rate is an upper bound on the mutual information, which is...
UAI

Learning proposals for probabilistic programs with inference combinators

Stites, Sam, Zimmermann, Heiko, Wu, Hao, Sennesh, Eli, and Meent, Jan-Willem

In Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence 27–30 jul 2021

Abs HTML PDF Code

We develop operators for construction of proposals in probabilistic programs, which we refer to as inference combinators. Inference combinators define a grammar over importance samplers that compose primitive operations such as application of a transition kernel and importance resampling. Proposals in these samplers can be parameterized using neural networks, which in turn can be trained by optimizing variational objectives. The result is a framework for user-programmable variational methods that are correct by construction and can be tailored to specific models. We demonstrate the flexibility of this framework by implementing advanced variational methods based on amortized Gibbs sampling and annealing.
EMNLP

Disentangling Representations of Text by Masking Transformers

Zhang, Xiongyi, van de Meent, Jan-Willem, and Wallace, Byron

In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing Nov 2021

Abs HTML PDF

Representations from large pretrained models such as BERT encode a range of features into monolithic vectors, affording strong predictive accuracy across a range of downstream tasks. In this paper we explore whether it is possible to learn disentangled representations by identifying existing subnetworks within pretrained models that encode distinct, complementary aspects. Concretely, we learn binary masks over transformer weights or hidden units to uncover subsets of features that correlate with a specific factor of variation; this eliminates the need to train a disentangled model from scratch for a particular task. We evaluate this method with respect to its ability to disentangle representations of sentiment from genre in movie reviews, toxicity from dialect in Tweets, and syntax from semantics. By combining masking with magnitude pruning we find that we can identify sparse subnetworks within BERT that strongly encode particular aspects (e.g., semantics) while only weakly encoding others (e.g., syntax). Moreover, despite only learning masks, disentanglement-via-masking performs as well as — and often better than —previously proposed methods based on variational autoencoders and adversarial training.
NAACL

On the Impact of Random Seeds on the Fairness of Clinical Classifiers

Amir, Silvio, van de Meent, Jan-Willem, and Wallace, Byron

In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies Jun 2021

Abs HTML PDF

Recent work has shown that fine-tuning large networks is surprisingly sensitive to changes in random seed(s). We explore the implications of this phenomenon for model fairness across demographic groups in clinical prediction tasks over electronic health records (EHR) in MIMIC-III —— the standard dataset in clinical NLP research. Apparent subgroup performance varies substantially for seeds that yield similar overall performance, although there is no evidence of a trade-off between overall and subgroup performance. However, we also find that the small sample sizes inherent to looking at intersections of minority groups and somewhat rare conditions limit our ability to accurately estimate disparities. Further, we find that jointly optimizing for high overall performance and low disparities does not yield statistically significant improvements. Our results suggest that fairness work using MIMIC-III should carefully account for variations in apparent differences that may arise from stochasticity and small sample sizes.
NeurIPS

Nested Variational Inference

Zimmermann, Heiko, Wu, Hao, Esmaeili, Babak, and Meent, Jan-Willem

In Advances in Neural Information Processing Systems Jun 2021

Abs HTML PDF

We develop nested variational inference (NVI), a family of methods that learn proposals for nested importance samplers by minimizing an forward or reverse KL divergence at each level of nesting. NVI is applicable to many commonly-used importance sampling strategies and provides a mechanism for learning intermediate densities, which can serve as heuristics to guide the sampler. Our experiments apply NVI to (a) sample from a multimodal distribution using a learned annealing path (b) learn heuristics that approximate the likelihood of future observations in a hidden Markov model and (c) to perform amortized inference in hierarchical deep generative models. We observe that optimizing nested objectives leads to improved sample quality in terms of log average weight and effective sample size.
AAMAS

Action Priors for Large Action Spaces in Robotics

Biza, Ondrej, Wang, Dian, Platt, Robert, Meent, Jan-Willem, and Wong, Lawson L.S.

In Proceedings of the 20th International Conference on Autonomous Agents and MultiAgent Systems Jun 2021

Abs HTML PDF Code

In robotics, it is often not possible to learn useful policies using pure model-free reinforcement learning without significant reward shaping or curriculum learning. As a consequence, many researchers rely on expert demonstrations to guide learning. However, acquiring expert demonstrations can be expensive. This paper proposes an alternative approach where the solutions of previously solved tasks are used to produce an action prior that can facilitate exploration in future tasks. The action prior is a probability distribution over actions that summarizes the set of policies found solving previous tasks. Our results indicate that this approach can be used to solve robotic manipulation problems that would otherwise be infeasible without expert demonstrations. Source code is available at https://github.com/ondrejba/action_priors.
ICML

Conjugate Energy-Based Models

Wu*, Hao, Esmaeili*, Babak, Wick, Michael L, Tristan, Jean-Baptiste, and van de Meent, Jan-Willem

In International Conference on Machine Learning Jun 2021

Abs

In this paper, we propose conjugate energy-based models (CEBMs), a new class of energy-based models that define a joint density over data and latent variables. The joint density of a CEBM decomposes into an intractable distribution over data and a tractable posterior over latent variables. CEBMs have similar use cases as variational autoencoders, in the sense that they learn an unsupervised mapping from data to latent variables. However, these models omit a generator network, which allows them to learn more flexible notions of similarity between data points. Our experiments demonstrate that conjugate EBMs achieve competitive results in terms of image modelling, predictive power of latent space, and out-of-domain detection on a variety of datasets.