Guy Wolf

Biographie

Guy Wolf est professeur titulaire au Département de mathématiques et de statistique (DMS) de l'Université de Montréal (UdeM), titulaire d'une chaire en IA Canada-CIFAR et membre académique principal de Mila (l'Institut québécois d'intelligence artificielle), chercheur associé au CRCHUM (Centre de recherche du Centre hospitalier de l'Université de Montréal) et chercheur principal participant au Laboratoire international Helmholtz pour la dynamique cellulaire causale.

En 2024, il a reçu une bourse de recherche Humboldt pour chercheurs expérimentés, dans le cadre de laquelle il a été professeur invité à l'Université de Heidelberg (2024) et à Helmholtz Munich (2024-2026) en Allemagne. Avant de joindre l'UdeM et Mila, il a été professeur adjoint Gibbs (2015-2018) au sein du programme de mathématiques appliquées, puis chercheur scientifique associé au Département de génétique (2018) de l'Université Yale (Connecticut, États-Unis). Auparavant, il a travaillé comme chercheur postdoctoral (2013-2015) au Département d'informatique de l'École normale supérieure à Paris (France). Il détient un doctorat en informatique de l'Université de Tel-Aviv (Israel) et possède cinq ans d'expérience préalable en conception et développement de logiciels informatiques pour l'analyse de données en contexte militaire.

Ses recherches actuelles portent sur l'apprentissage guidé de représentations pour l'exploration de données, notamment par des méthodes qui exploitent l'apprentissage de variétés (manifold learning) et l'apprentissage profond géométrique pour la réduction de dimensionnalité, la visualisation, le débruitage, l'augmentation de données et la modélisation à gros grains (coarse graining). Bien que ces approches s'appliquent à un large éventail de domaines, il s'intéresse particulièrement à l'intersection de l'IA et de la santé, notamment aux outils facilitant l'analyse exploratoire de données biomédicales, comme dans les domaines de la multiomique sur cellule unique (single-cell multiomics), de la découverte de médicaments et des neurosciences.

Étudiants actuels

Adrien Aumon

Doctorat - UdeM

adrien.aumon@mila.quebec

Semih Cantürk

Doctorat - UdeM

semihcanturk00@gmail.com

Joao Felipe Carneiro Barbosa Rocha

Collaborateur·rice de recherche - Yale University

Co-superviseur⋅e :

Smita Krishnaswamy

joaofelipe.rocha@mila.quebec

sebastian.damrich@mila.quebec

Sebastian Damrich

Collaborateur·rice de recherche - University of Tübingen

Maxine Dussault

Maîtrise recherche - UdeM

Co-superviseur⋅e :

Margarida Carvalho

maxine.dussault@mila.quebec

Stefan Horoi

Doctorat - UdeM

stefan.horoi@mila.quebec

Xiao Huang

Maîtrise recherche - Concordia

Superviseur⋅e principal⋅e :

xiaolong.huang@mila.quebec

Collaborateur·rice alumni - UdeM

guillaume.huguet@mila.quebec

Paul Janson

Doctorat - Concordia

Superviseur⋅e principal⋅e :

myriam.lizotte@mila.quebec

paul.janson@mila.quebec

Doctorat - UdeM

Gabriele Malagoli

Visiteur de recherche indépendant - Helmholtz Munich

gabriele.malagoli@mila.quebec

Philippe Martin

Doctorat - UdeM

Co-superviseur⋅e :

Paul François

philippe.martin@mila.quebec

Paria Mehrbod

Maîtrise recherche - Concordia

Superviseur⋅e principal⋅e :

paria.mehrbod@mila.quebec

Lydia Mezrag

Doctorat - UdeM

lydia.mezrag@mila.quebec

Kevin Moon

Collaborateur·rice de recherche

Sacha Morin

Doctorat - UdeM

Co-superviseur⋅e :

Liam Paull

sacha.morin@mila.quebec

Postdoctorat - Concordia

Superviseur⋅e principal⋅e :

geraldin.nanfack@mila.quebec

Shuang Ni

Doctorat - UdeM

shuang.ni@mila.quebec

Albert Orozco Camacho

Doctorat - Concordia

Superviseur⋅e principal⋅e :

Jake Rhodes

Collaborateur·rice de recherche - BYU

rhodes@stat.byu.edu

thomas.sabourin@mila.quebec

Thomas Sabourin

Maîtrise recherche - UdeM

johannes.schmidt@mila.quebec

Johannes Schmidt

Visiteur de recherche indépendant - University of Fribourg

Doctorat - UdeM

Superviseur⋅e principal⋅e :

Julie Hussin

matthew.scicluna@mila.quebec

francisco.jose.tellez@mila.quebec

Simeon Tchoffo

Doctorat - UdeM

simeontchoffo@gmail.com

Francisco Tellez

Maîtrise recherche - UdeM

Jesuino Vieira Filho

Maîtrise recherche - UdeM

Collaborateur·rice alumni - UdeM

Co-superviseur⋅e :

yanlei.zhang@mila.quebec

Analyser le paradoxe des interférons inhérent à la COVID-19 au moyen de la réduction de la dimensionnalité et du regroupement

Billets de blogue

Graph and representation of working methodology, and graph of data on deaths 60 days after onset of symptoms.

19 février 2025

par

Sacha Morin

Elsa Brunet-Ratnasingham

Guy Wolf

Lire l'article

Publications

Freeze, Diffuse, Decode: Task-Aware Adaptation of Transformer Embeddings for Antimicrobial Peptide Design

Pankhil Gawade

Adam Izdebski

Myriam Lizotte

Kevin R. Moon

Jake Slater Rhodes

Ewa Szczurek

Pretrained transformers provide general-purpose molecular embeddings for downstream tasks. While these embeddings provide task-agnostic stru… (voir plus)ctural patterns, they lack task-specific alignment, limiting downstream performance. Here, we introduce Freeze, Diffuse, Decode (FDD), a diffusion-based framework that adapts pretrained transformer embeddings to downstream tasks by building a task supervised diffusion geometry over the frozen embeddings, without any backbone training. Applied to antimicrobial peptide design, FDD yields low-dimensional, predictive, and interpretable representations that support property prediction, retrieval, and latent-space interpolation.

2026-07-15

Topology, Algebra, and Geometry in Data Science (présentation orale)

César Miguel Valdez Cordova

Towards Distilling the Representational Geometry of Language Models

Andrew J Steindl

Dhananjay Bhaskar

Ian Adelstein

Smita Krishnaswamy

Large neural networks are increasingly combined through weight-space merging and distilled into smaller models, but weights and activations … (voir plus)are not canonical coordinates: two networks can implement similar functions while differing by permutations, rescalings, or basis changes. We ask whether these operations can instead be guided by the geometry of a model's representation space. For each model, we summarize the relationships among probe examples using a diffusion operator built from internal activations. We then test a simple distillation procedure in which a student is trained to match a fixed diffusion-geometry target derived from a teacher. Across Pythia, BERT, and DeBERTa-v3, this target does not generally make the student inherit the teacher's specific representation geometry. Instead, distilled students usually remain closer to matched task-only controls, and the objective has little effect on held-out performance. These results suggest that representation-geometry objectives must be evaluated against the geometry a student would learn from ordinary training alone.

2026-07-15

Topology, Algebra, and Geometry in Data Science (poster)

Can Model Merging Improve Aggregation in DiLoCo?

Model merging techniques, which aggregate independently finetuned models into one to combine their capabilities, have become a topic of sign… (voir plus)ificant interest in recent years, with a broad array of methods having been proposed to tackle this problem. Simultaneously, an emerging trend in distributed learning has been the use of methods such as local SGD and DiLoCo, which greatly reduce communication costs by periodically aggregating the independently trained local models. However, these communication-efficient methods have been shown to degrade in performance relative to the FLOP-matched data-parallel gold standard as the number of independent local models grows and as the number of local training steps before global communication is increased. In this work, we draw an explicit analogy between the pseudo-gradient aggregation step in local SGD/DiLoCo and task arithmetic-based model merging, establishing a straightforward way to utilize merging methods in the context of distributed optimization. We then evaluate multiple state-of-the-art model merging methods in this setting and identify one method in particular, Iso-C, as a promising approach for improving DiLoCo. We find that DiLoCo SGD with Iso-C aggregation outperforms not only simple pseudo-gradient averaging but even the momentum-based DiLoCo, despite lacking a momentum mechanism itself. Building on this finding, we propose IsoLoCo, which adapts Iso-C for distributed training by equipping it with Nesterov momentum. Our empirical evaluations on language model pre-training across varying numbers of local workers show that IsoLoCo significantly outperforms DiLoCo, with the gap between them widening as the number of workers increases. This advantage remains present across model sizes and inner step counts, confirming that merging-inspired aggregation is an effective strategy for low-communication distributed training.

2026-07-02

arXiv (prépublication)

Gaining Biological Insights through Supervised Data Visualization

Jake S. Rhodes

Adrien Aumon

Sacha Morin

Marc Girard

Catherine Larochelle

Boaz Lahav

Elsa Brunet-Ratnasingham

Wei Zhang

Adele Cutler

Anhong Zhou

Daniel E. Kaufmann

Stephanie Zandee

Alexandre Prat

Kevin R. Moon

Dimensionality reduction-based data visualization is pivotal in comprehending complex biological data. The most common methods, such as PHAT… (voir plus)E, t-SNE, and UMAP, are unsupervised and therefore reflect the dominant structure in the data, which may be independent of expert-provided labels. Here we introduce a supervised data visualization method called RF-PHATE, which integrates expert knowledge for further exploration of the data. RF-PHATE leverages random forests to capture intricate featurelabel relationships. Extracting information from the forest, RF-PHATE generates low-dimensional visualizations that highlight relevant data relationships while disregarding extraneous features. This approach scales to large datasets and applies to classification and regression. We illustrate RF-PHATE’s prowess through three case studies. In a multiple sclerosis study using longitudinal clinical and imaging data, RF-PHATE unveils a sub-group of patients with non-benign relapsingremitting Multiple Sclerosis, demonstrating its aptitude for time-series data. In the context of Raman spectral data, RF-PHATE effectively showcases the impact of antioxidants on diesel exhaust-exposed lung cells, highlighting its proficiency in noisy environments. Furthermore, RF-PHATE aligns established geometric structures with COVID-19 patient outcomes, enriching interpretability in a hierarchical manner. RF-PHATE bridges expert insights and visualizations, promising knowledge generation. Its adaptability, scalability, and noise tolerance underscore its potential for widespread adoption.

2026-06-29

Nature Computational Science (publié)

Path-independent Flow Matching for Multi-parameter Generative Dynamics

Sina Sanjari

Yanlei Zhang

Flow Matching is a powerful framework for learning transport maps between probability distributions. Yet its standard single-parameter formu… (voir plus)lation is not designed to capture multi-parameter variations where the resulting transport should be path-independent. Path independence is crucial because it ensures that transformations depend only on the initial and target distributions, not on the specific path. In this work, we introduce Path-independent Flow Matching (PiFM), a method for learning vector fields whose induced flows yield path-independent transport between distributions. We show that PiFM generalizes Flow Matching to higher-dimensional parameter domains while enforcing structural conditions that ensure consistency of composed transformations. In addition, we show that, under suitable assumptions, PiFM approximates the Wasserstein barycenter, linking the framework to a notion of distributional interpolation. To enable practical training, we propose a tractable, simulation-free objective that regresses onto multi-parameter conditional probability paths. We showcase empirically that PiFM outperforms other approaches on both synthetic and real world data in interpolating path-independent trajectories and generating desired out of distribution samples.

2026-05-29

SPIGM @ International Conference on Machine Learning (poster)

Meta-Merging by Checkpoint Nowcasting

Albert Manuel Orozco Camacho

Boris Knyazev

Model merging---the direct combination of parameters from independently fine-tuned networks---offers a way to compose task-specific capabili… (voir plus)ties without retraining or ensemble inference. Existing merge methods are often built from hand-crafted arithmetic or sparsification heuristics, leaving open whether general learned weight-space operators can be repurposed for merging directly. We study this question with NiNo, a pre-trained checkpoint-nowcasting meta-network originally designed to predict near-future training states from short checkpoint histories. We show that pre-trained NiNo can be reused as a data-free pairwise meta-merge operator for independently fine-tuned models. On an 8-task CLIP ViT-B/16 benchmark, NiNo is competitive with strong arithmetic baselines and consistently lands in the same functional region as weight averaging, Task Arithmetic, and TIES. Moreover, NiNo is best on HumanEval in a Qwen3 language extension among the compared merge methods, while extending meta-merge beyond pairs remains an open challenge. These results position learned checkpoint nowcasting as a practical starting point for data-free model merging and motivate future weight-space learners trained for merging explicitly.

2026-05-23

WSS @ International Conference on Machine Learning (poster)

César Miguel Valdez Cordova

scShapeBench: Discovering geometry from high dimensional scRNAseq data

Andrew J. Steindl

João Felipe Rocha

Brian Tshilengi Di Bassinga

Zachary Warren

Matthew Scicluna

Shabarni Gupta

Leire Torices

Daniel Neumann

Timothy J. Mann

Ihuan Gunawan

Dhananjay Bhaskar

John G. Lock

Christine L. Chaffer

Smita Krishnaswamy

High-dimensional point cloud data arise across many scientific domains, especially single-cell biology. The shapes or topologies of these da… (voir plus)tasets determine the types of information that can be extracted. For example, clustered data supports cell-type identification, trajectory structures support transition analysis, and archetypal structures capture continua of cellular behaviors. Existing analysis pipelines often assume a specific shape. The standard Seurat pipeline combines UMAP visualization with Louvain clustering and therefore assumes clustered data, while tools such as Monocle and SPADE assume tree-like structures, and flow-based models such as MIOFlow and Conditional Flow Matching target trajectories. Choosing which pipeline to apply is therefore often left to bioinformaticians who visually inspect datasets before selecting an analysis strategy. With the rise of agentic AI scientists, automating shape detection is increasingly important for selecting downstream analysis pipelines. To address this problem, we introduce scShapeBench, a benchmark dataset for shape detection containing both synthetic and expert-annotated single-cell datasets. Synthetic datasets are sampled from ground-truth skeleton graphs with controlled variance. Real single-cell datasets are curated from diverse sources and annotated by experts into four categories: clusters, single trajectory, multi-branching, and archetypal. We additionally introduce scReebTower, a baseline method that uses diffusion geometry to extract Reeb graphs and connect visualization with pipeline selection. We provide topology-aware evaluation metrics and compare scReebTower against PAGA and Mapper on synthetic and real data. Our results indicate that scReebTower outperforms existing baselines. Overall, our contributions span benchmarks, evaluation metrics, and a baseline for automated shape detection in single-cell data.

2026-05-11

arXiv (prépublication)

Diversity Curves for Graph Representation Learning

Katharina Limbeck

Nadja Häusermann

Martin Carrasco

Bastian Rieck

Graph-level representations are crucial tools for characterising structural differences between graphs. However, comparing graphs with diffe… (voir plus)rent cardinalities, even when sampled from the same underlying distribution, remains challenging. Unsupervised tasks in particular require interpretable, scalable, and reliable size-aware graph representations. Our work addresses these issues by tracking the structural diversity of a graph across coarsening levels. The resulting graph embeddings, which we denote diversity curves, are interpretable by construction, efficient, and directly comparable across coarsening hierarchies. Specifically, we track the spread of graphs, a novel isometry invariant that is inherently well-suited for encoding the metric diversity and geometry of graphs. We utilise edge contraction coarsening and prove that this improves expressivity, thus leading to more powerful graph-level representations than structural descriptors alone. Demonstrating their utility over a range of baseline methods in practice, we use diversity curves to (i) cluster and visualise simulated graphs across varying sizes, (ii) distinguish the geometry of single-cell graphs, (iii) compare the structure of molecular graph datasets, and (iv) characterise geometric shapes.

2026-05-06

arXiv (prépublication)

No Triangulation Without Representation: Generalization in Topological Deep Learning

Johannes S. Schmidt

Martin Carrasco

Ernst Röell

Nello Blaser

Bastian Rieck

Despite an ever-increasing interest in topological deep learning models that target higher-order datasets, there is no consensus on how to e… (voir plus)valuate such models. This is exacerbated by the fact that topological objects permit operations, such as structural refinements, that are not appropriate for graph data. In this work, we extend MANTRA, a benchmark dataset containing manifold triangulations, to a larger class of manifolds with more diverse homeomorphism types. We show that, unlike prior claims, both graph neural networks (GNNs) and higher-order message passing (HOMP) methods can saturate the benchmark. However, we find that this is contingent on the right representation and feature assignment, emphasizing their importance in baseline models. We thus provide a novel evaluation protocol based on representational diversity and triangulation refinement. Surprisingly, we find no indication that existing models are capable of generalizing beyond the combinatorial structure of the data. This points towards a research gap in developing models that understand topological structure independent of scale. Our work thus provides the necessary scaffolding to evaluate future models and enable the development of topology-aware inductive biases.

2026-05-06

arXiv (prépublication)

Can Computational Reducibility Lead to Transferable Models for Graph Combinatorial Optimization?

Michael Perlmutter

A key challenge in deriving unified neural solvers for combinatorial optimization (CO) is efficient generalization of models between a given… (voir plus) set of tasks to new tasks not used during the initial training process. To address it, we first establish a new model, which uses a GCON module as a form of expressive message passing together with energy-based unsupervised loss functions. This model achieves high performance (often comparable with state-of-the-art results) across multiple CO tasks when trained individually on each task. We then leverage knowledge from the computational reducibility literature to propose pretraining and fine-tuning strategies that transfer effectively (a) between MVC, MIS and MaxClique, and (b) in a multi-task learning setting that additionally incorporates MaxCut, MDS and graph coloring. Additionally, in a leave-one-out, multi-task learning setting, we observe that pretraining on all but one task almost always leads to faster convergence on the remaining task when fine-tuning while avoiding negative transfer. Our findings indicate that learning common representations across multiple graph CO problems is viable through the use of expressive message passing coupled with pretraining strategies that are informed by the polynomial reduction literature, thereby taking an important step towards enabling the development of foundational models for neural CO. We provide an open-source implementation of our work at https://github.com/semihcanturk/COPT-MT .

2026-04-29

International Conference on Machine Learning (accepté)

Geometry-aware graph attention networks to explain single-cell chromatin states and gene expression with SEAGALL

Gabriele Malagoli

Patrick Hanel

Anna Danese

Maria Colomé-Tatché

High-throughput single-cell sequencing is widely used to study cell identity. We present SEAGALL (Single-cell Explainable Geometry-Aware Gra… (voir plus)ph Attention Learning pipeLine), a deep learning method to quantify the impact of molecular features on cellular phenotype, based on geometry-regularised autoencoders (GRAE) and explainable graph attention networks (X-GAT). The GRAE embeds the data into a latent space to build a reliable cell-cell graph. The GAT is trained to learn the annotations and XAI is used to explain the predictions, unravelling the features driving cell identity. SEAGALL extracts specific and stable signatures from multiple omics experiments, going beyond differential marker genes.

2026-04-22

Genome Biology (publié)

Active search generation for nanophotonic design in the small data regime

Vincent Létourneau

Yuri Grinberg

Dan Kushnir

Yanlei Zhang

Dan-Xia Xu

2026-04-09

Machine Learning in Photonics (publié)