Author name cluster

Pierre Ablin

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

28 papers

2 author rows

ICML Conference 2025 Conference Paper

Scaling Laws for Forgetting during Finetuning with Pretraining Data Injection

Louis Béthune
David Grangier
Dan Busbridge
Eleonora Gualdoni
Marco Cuturi
Pierre Ablin

A widespread strategy to obtain a language model that performs well on a target domain is to finetune a pretrained model to perform unsupervised next-token prediction on data from that target domain. Finetuning presents two challenges: (i) if the amount of target data is limited, as in most practical applications, the model will quickly overfit, and (ii) the model will drift away from the original model, forgetting the pretraining data and the generic knowledge that comes with it. Our goal is to derive scaling laws that quantify these two phenomena for various target domains, amounts of available target data, and model scales. We measure the efficiency of injecting pretraining data into the finetuning data mixture to avoid forgetting and mitigate overfitting. A key practical takeaway from our study is that injecting as little as $1%$ of pretraining data in the finetuning data mixture prevents the model from forgetting the pretraining set.

Details

NeurIPS Conference 2025 Conference Paper

Scaling Laws for Optimal Data Mixtures

Mustafa Shukor
Louis Bethune
Dan Busbridge
David Grangier
Enrico Fini
Alaaeldin El-Nouby
Pierre Ablin

Large foundation models are typically trained on data from multiple domains, with the data mixture--the proportion of each domain used--playing a critical role in model performance. The standard approach to selecting this mixture relies on trial and error, which becomes impractical for large-scale pretraining. We propose a systematic method to determine the optimal data mixture for any target domain using scaling laws. Our approach accurately predicts the loss of a model of size $N$ trained with $D$ tokens and a specific domain weight vector $h$. We validate the universality of these scaling laws by demonstrating their predictive power in three distinct and large-scale settings: large language model (LLM), native multimodal model (NMM), and large vision models (LVM) pretraining. We further show that these scaling laws can extrapolate to new data mixtures and across scales: their parameters can be accurately estimated using a few small-scale training runs, and used to estimate the performance at larger scales and unseen domain weights. The scaling laws allow to derive the optimal domain weights for any target domain under a given training budget ($N$, $D$), providing a principled alternative to costly trial-and-error methods.

PDF Details

ICML Conference 2025 Conference Paper

Shielded Diffusion: Generating Novel and Diverse Images using Sparse Repellency

Michael Kirchhof 0002
James Thornton
Louis Béthune
Pierre Ablin
Eugène Ndiaye
Marco Cuturi

The adoption of text-to-image diffusion models raises concerns over reliability, drawing scrutiny under the lens of various metrics like calibration, fairness, or compute efficiency. We focus in this work on two issues that arise when deploying these models: a lack of diversity when prompting images, and a tendency to recreate images from the training set. To solve both problems, we propose a method that coaxes the sampled trajectories of pretrained diffusion models to land on images that fall outside of a reference set. We achieve this by adding repellency terms to the diffusion SDE throughout the generation trajectory, which are triggered whenever the path is expected to land too closely to an image in the shielded reference set. Our method is sparse in the sense that these repellency terms are zero and inactive most of the time, and even more so towards the end of the generation trajectory. Our method, named SPELL for sparse repellency, can be used either with a static reference set that contains protected images, or dynamically, by updating the set at each timestep with the expected images concurrently generated within a batch, and with the images of previously generated batches. We show that adding SPELL to popular diffusion models improves their diversity while impacting their FID only marginally, and performs comparatively better than other recent training-free diversity methods. We also demonstrate how SPELL can ensure a shielded generation away from a very large set of protected images by considering all 1. 2M images from ImageNet as the protected set.

Details

ICML Conference 2025 Conference Paper

Soup-of-Experts: Pretraining Specialist Models via Parameters Averaging

Pierre Ablin
Angelos Katharopoulos
Skyler Seto
David Grangier

Machine learning models are routinely trained on a mixture of different data domains. Different domain weights yield very different downstream performances. We propose the Soup-of-Experts, a novel architecture that can instantiate a model at test time for any domain weights with minimal computational cost and without re-training the model. Our architecture consists of a bank of expert parameters, which are linearly combined to instantiate one model. We learn the linear combination coefficients as a function of the input domain weights. To train this architecture, we sample random domain weights, instantiate the corresponding model, and backprop through one batch of data sampled with these domain weights. We demonstrate how our approach obtains small specialized models on several language modeling tasks quickly. Soup-of-Experts are particularly appealing when one needs to ship many different specialist models quickly under a size constraint.

Details

ICLR Conference 2025 Conference Paper

Task-Adaptive Pretrained Language Models via Clustered-Importance Sampling

David Grangier
Simin Fan
Skyler Seto
Pierre Ablin

Specialist language models (LMs) focus on a specific task or domain on which they often outperform generalist LMs of the same size. However, the specialist data needed to pretrain these models is only available in limited amount for most tasks. In this work, we build specialist models from large generalist training sets instead. We adjust the training distribution of the generalist data with guidance from the limited domain-specific data. We explore several approaches, with clustered importance sampling standing out. This method clusters the generalist dataset and samples from these clusters based on their frequencies in the smaller specialist dataset. It is scalable, suitable for pretraining and continued pretraining, it works well in multi-task settings. Our findings demonstrate improvements across different domains in terms of language modeling perplexity and accuracy on multiple-choice question tasks. We also present ablation studies that examine the impact of dataset sizes, clustering configurations, and model sizes.

Details

ICLR Conference 2025 Conference Paper

The AdEMAMix Optimizer: Better, Faster, Older

Matteo Pagliardini
Pierre Ablin
David Grangier

Momentum based optimizers are central to a wide range of machine learning applications. These typically rely on an Exponential Moving Average (EMA) of gradients, which decays exponentially the present contribution of older gradients. This accounts for gradients being local linear approximations which lose their relevance as the iterate moves along the loss landscape. This work questions the use of a single EMA to accumulate past gradients and empirically demonstrates how this choice can be sub-optimal: a single EMA cannot simultaneously give a high weight to the immediate past, and a non-negligible weight to older gradients. Building on this observation, we propose AdEMAMix, a simple modification of the Adam optimizer with a mixture of two EMAs to better take advantage of past gradients. Our experiments on language modeling and image classification show---quite surprisingly---that gradients can stay relevant for tens of thousands of steps. They help to converge faster, and often to lower minima: e.g., a $1.3$B parameter AdEMAMix LLM trained on $101$B tokens performs comparably to an AdamW model trained on $197$B tokens ($+95\%$). Moreover, our method significantly slows-down model forgetting during training. Our work motivates further exploration of different types of functions to leverage past gradients, beyond EMAs.

Details

ICLR Conference 2025 Conference Paper

Theory, Analysis, and Best Practices for Sigmoid Self-Attention

Jason Ramapuram
Federico Danieli
Eeshan Gunesh Dhekane
Floris Weers
Dan Busbridge
Pierre Ablin
Tatiana Likhomanenko
Jagrit Digani

Attention is a key part of the transformer architecture. It is a sequence-to-sequence mapping that transforms each sequence element into a weighted sum of values. The weights are typically obtained as the softmax of dot products between keys and queries. Recent work has explored alternatives to softmax attention in transformers, such as ReLU and sigmoid activations. In this work, we revisit sigmoid attention and conduct an in-depth theoretical and empirical analysis. Theoretically, we prove that transformers with sigmoid attention are universal function approximators and benefit from improved regularity compared to softmax attention. Through detailed empirical analysis, we identify stabilization of large initial attention norms during the early stages of training as a crucial factor for the successful training of models with sigmoid attention, outperforming prior attempts. We also introduce FLASHSIGMOID, a hardware-aware and memory-efficient implementation of sigmoid attention yielding a 17% inference kernel speed-up over FLASHATTENTION2 on H100 GPUs. Experiments across language, vision, and speech show that properly normalized sigmoid attention matches the strong performance of softmax attention on a wide range of domains and scales, which previous attempts at sigmoid attention were unable to fully achieve. Our work unifies prior art and establishes best practices for sigmoid attention as a drop-in softmax replacement in transformers.

Details

TMLR Journal 2024 Journal Article

Adaptive Training Distributions with Scalable Online Bilevel Optimization

David Grangier
Pierre Ablin
Awni Hannun

Large neural networks pretrained on web-scale corpora are central to modern machine learning. In this paradigm, the distribution of the large, heterogeneous pretraining data rarely matches that of the application domain. This work considers modifying the pretraining distribution in the case where one has a small sample of data reflecting the targeted test conditions. We propose an algorithm motivated by a recent formulation of this setting as an online, bilevel optimization problem. With scalability in mind, our algorithm prioritizes computing gradients at training points which are likely to most improve the loss on the targeted distribution. Empirically, we show that in some cases this approach is beneficial over existing strategies from the domain adaptation literature but may not succeed in other cases. We propose a simple test to evaluate when our approach can be expected to work well and point towards further research to address current limitations.

PDF Details

ICML Conference 2024 Conference Paper

Careful with that Scalpel: Improving Gradient Surgery with an EMA

Yu-Guan Hsieh
James Thornton
Eugène Ndiaye
Michal Klein
Marco Cuturi
Pierre Ablin

Beyond minimizing a single training loss, many deep learning estimation pipelines rely on an auxiliary objective to quantify and encourage desirable properties of the model (e. g. performance on another dataset, robustness, agreement with a prior). Although the simplest approach to incorporating an auxiliary loss is to sum it with the training loss as a regularizer, recent works have shown that one can improve performance by blending the gradients beyond a simple sum; this is known as gradient surgery. We cast the problem as a constrained minimization problem where the auxiliary objective is minimized among the set of minimizers of the training loss. To solve this bilevel problem, we follow a parameter update direction that combines the training loss gradient and the orthogonal projection of the auxiliary gradient to the training gradient. In a setting where gradients come from mini-batches, we explain how, using a moving average of the training loss gradients, we can carefully maintain this critical orthogonality property. We demonstrate that our method, Bloop, can lead to much better performances on NLP and vision experiments than other gradient surgery methods without EMA.

Details

ICML Conference 2024 Conference Paper

How Smooth Is Attention?

Valérie Castin
Pierre Ablin
Gabriel Peyré

Self-attention and masked self-attention are at the heart of Transformers’ outstanding success. Still, our mathematical understanding of attention, in particular of its Lipschitz properties — which are key when it comes to analyzing robustness and expressive power — is incomplete. We provide a detailed study of the Lipschitz constant of self-attention in several practical scenarios, discussing the impact of the sequence length $n$ and layer normalization on the local Lipschitz constant of both unmasked and masked self-attention. In particular, we show that for inputs of length $n$ in any compact set, the Lipschitz constant of self-attention is bounded by $\sqrt{n}$ up to a constant factor and that this bound is tight for reasonable sequence lengths. When the sequence length $n$ is too large for the previous bound to be tight, which we refer to as the mean-field regime, we provide an upper bound and a matching lower bound which are independent of $n$. Our mean-field framework for masked self-attention is novel and of independent interest. Our experiments on pretrained and randomly initialized BERT and GPT-2 support our theoretical findings.

Details

JMLR Journal 2024 Journal Article

Infeasible Deterministic, Stochastic, and Variance-Reduction Algorithms for Optimization under Orthogonality Constraints

Pierre Ablin
Simon Vary
Bin Gao
Pierre-Antoine Absil

Orthogonality constraints naturally appear in many machine learning problems, from principal component analysis to robust neural network training. They are usually solved using Riemannian optimization algorithms, which minimize the objective function while enforcing the constraint. However, enforcing the orthogonality constraint can be the most time-consuming operation in such algorithms. Recently, Ablin and Peyré (2022) proposed the landing algorithm, a method with cheap iterations that does not enforce the orthogonality constraints but is attracted towards the manifold in a smooth manner. This article provides new practical and theoretical developments for the landing algorithm. First, the method is extended to the Stiefel manifold, the set of rectangular orthogonal matrices. We also consider stochastic and variance reduction algorithms when the cost function is an average of many functions. We demonstrate that all these methods have the same rate of convergence as their Riemannian counterparts that exactly enforce the constraint, and converge to the manifold. Finally, our experiments demonstrate the promise of our approach to an array of machine-learning problems that involve orthogonality constraints. [abs] [ pdf ][ bib ] [ code ] &copy JMLR 2024. ( edit, beta )

PDF Details

NeurIPS Conference 2024 Conference Paper

Learning Elastic Costs to Shape Monge Displacements

Michal Klein
Aram-Alexandre Pooladian
Pierre Ablin
Eugène Ndiaye
Jonathan Niles-Weed
Marco Cuturi

Given a source and a target probability measure, the Monge problem studies efficient ways to map the former onto the latter. This efficiency is quantified by defining a *cost* function between source and target data. Such a cost is often set by default in the machine learning literature to the squared-Euclidean distance, $\ell^2\_2(\mathbf{x}, \mathbf{y}): =\tfrac12\|\mathbf{x}-\mathbf{y}\|\_2^2$. The benefits of using *elastic* costs, defined using a regularizer $\tau$ as $c(\mathbf{x}, \mathbf{y}): =\ell^2_2(\mathbf{x}, \mathbf{y})+\tau(\mathbf{x}-\mathbf{y})$, was recently highlighted in (Cuturi et al. 2023). Such costs shape the *displacements* of Monge maps $T$, namely the difference between a source point and its image $T(\mathbf{x})-\mathbf{x}$, by giving them a structure that matches that of the proximal operator of $\tau$. In this work, we make two important contributions to the study of elastic costs: *(i)* For any elastic cost, we propose a numerical method to compute Monge maps that are provably optimal. This provides a much-needed routine to create synthetic problems where the ground-truth OT map is known, by analogy to the Brenier theorem, which states that the gradient of any convex potential is always a valid Monge map for the $\ell_2^2$ cost; *(ii)* We propose a loss to *learn* the parameter $\theta$ of a parameterized regularizer $\tau_\theta$, and apply it in the case where $\tau_{A}({\bf z}): =\|A^\perp {\bf z}\|^2_2$. This regularizer promotes displacements that lie on a low-dimensional subspace of $\mathbb{R}^d$, spanned by the $p$ rows of $A\in\mathbb{R}^{p\times d}$. We illustrate the soundness of our procedure on synthetic data, generated using our first contribution, in which we show near-perfect recovery of $A$'s subspace using only samples. We demonstrate the applicability of this method by showing predictive improvements on single-cell data tasks.

PDF Details DOI

ICML Conference 2024 Conference Paper

Optimization without Retraction on the Random Generalized Stiefel Manifold

Simon Vary
Pierre Ablin
Bin Gao 0007
Pierre-Antoine Absil

Optimization over the set of matrices $X$ that satisfy $X^\top B X = I_p$, referred to as the generalized Stiefel manifold, appears in many applications involving sampled covariance matrices such as the canonical correlation analysis (CCA), independent component analysis (ICA), and the generalized eigenvalue problem (GEVP). Solving these problems is typically done by iterative methods that require a fully formed $B$. We propose a cheap stochastic iterative method that solves the optimization problem while having access only to a random estimates of $B$. Our method does not enforce the constraint in every iteration; instead, it produces iterations that converge to critical points on the generalized Stiefel manifold defined in expectation. The method has lower per-iteration cost, requires only matrix multiplications, and has the same convergence rates as its Riemannian optimization counterparts that require the full matrix $B$. Experiments demonstrate its effectiveness in various machine learning applications involving generalized orthogonality constraints, including CCA, ICA, and the GEVP.

Details

NeurIPS Conference 2023 Conference Paper

How to Scale Your EMA

Dan Busbridge
Jason Ramapuram
Pierre Ablin
Tatiana Likhomanenko
Eeshan Gunesh Dhekane
Xavier Suau Cuadros
Russell Webb

Preserving training dynamics across batch sizes is an important tool for practical machine learning as it enables the trade-off between batch size and wall-clock time. This trade-off is typically enabled by a scaling rule, for example, in stochastic gradient descent, one should scale the learning rate linearly with the batch size. Another important machine learning tool is the model EMA, a functional copy of a target model, whose parameters move towards those of its target model according to an Exponential Moving Average (EMA) at a rate parameterized by a momentum hyperparameter. This model EMA can improve the robustness and generalization of supervised learning, stabilize pseudo-labeling, and provide a learning signal for Self-Supervised Learning (SSL). Prior works have not considered the optimization of the model EMA when performing scaling, leading to different training dynamics across batch sizes and lower model performance. In this work, we provide a scaling rule for optimization in the presence of a model EMA and demonstrate the rule's validity across a range of architectures, optimizers, and data modalities. We also show the rule's validity where the model EMA contributes to the optimization of the target model, enabling us to train EMA-based pseudo-labeling and SSL methods at small and large batch sizes. For SSL, we enable training of BYOL up to batch size 24, 576 without sacrificing performance, a 6$\times$ wall-clock time reduction under idealized hardware settings.

PDF Details

ICML Conference 2023 Conference Paper

Monge, Bregman and Occam: Interpretable Optimal Transport in High-Dimensions with Feature-Sparse Maps

Marco Cuturi
Michal Klein
Pierre Ablin

Optimal transport (OT) theory focuses, among all maps $T: \mathbb{R}^d\rightarrow \mathbb{R}^d$ that can morph a probability measure $\mu$ onto another $\nu$, on those that are the “thriftiest”, i. e. such that the average cost $c(x, T(x))$ between $x$ and its image $T(x)$ is as small as possible. Many computational approaches have been proposed to estimate such Monge maps when $c$ is the squared-Euclidean distance, e. g. , using entropic maps [Pooladian+2021], or input convex neural networks [Makkuva+2020, Korotin+2020]. We propose a new research direction, that leverages a specific translation invariant cost $c(x, y): =h(x-y)$ inspired by the elastic net. Here, $h: =\tfrac{1}{2}\|\cdot\|_2^2+\tau(\cdot)$, where $\tau$ is a convex function. We highlight a surprising link tying together a generalized entropic map for $h$, Bregman centroids induced by $h$, and the proximal operator of $\tau$. We show how setting $\tau$ to be a sparsity-inducing norm results in the first application of Occam ’s razor to transport. These maps yield, mechanically, displacement vectors $\Delta(x): = T(x)-x$ that are sparse, with sparsity patterns that vary depending on $x$. We showcase the ability of our method to estimate meaningful OT maps for high-dimensional single-cell transcription data. We use our methods in the $34000$-d space of gene counts for cells, without using a prior dimensionality reduction, thus retaining the ability to interpret all displacements at the gene level.

Details

NeurIPS Conference 2022 Conference Paper

A framework for bilevel optimization that enables stochastic and global variance reduction algorithms

Mathieu Dagréou
Pierre Ablin
Samuel Vaiter
Thomas Moreau

Bilevel optimization, the problem of minimizing a value function which involves the arg-minimum of another function, appears in many areas of machine learning. In a large scale empirical risk minimization setting where the number of samples is huge, it is crucial to develop stochastic methods, which only use a few samples at a time to progress. However, computing the gradient of the value function involves solving a linear system, which makes it difficult to derive unbiased stochastic estimates. To overcome this problem we introduce a novel framework, in which the solution of the inner problem, the solution of the linear system, and the main variable evolve at the same time. These directions are written as a sum, making it straightforward to derive unbiased estimates. The simplicity of our approach allows us to develop global variance reduction algorithms, where the dynamics of all variables is subject to variance reduction. We demonstrate that SABA, an adaptation of the celebrated SAGA algorithm in our framework, has $O(\frac1T)$ convergence rate, and that it achieves linear convergence under Polyak-Lojasciewicz assumption. This is the first stochastic algorithm for bilevel optimization that verifies either of these properties. Numerical experiments validate the usefulness of our method.

PDF Details

NeurIPS Conference 2022 Conference Paper

Benchopt: Reproducible, efficient and collaborative optimization benchmarks

Thomas Moreau
Mathurin Massias
Alexandre Gramfort
Pierre Ablin
Pierre-Antoine Bannier
Benjamin Charlier
Mathieu Dagréou
Tom Dupre la Tour

Numerical validation is at the core of machine learning research as it allows us to assess the actual impact of new methods, and to confirm the agreement between theory and practice. Yet, the rapid development of the field poses several challenges: researchers are confronted with a profusion of methods to compare, limited transparency and consensus on best practices, as well as tedious re-implementation work. As a result, validation is often very partial, which can lead to wrong conclusions that slow down the progress of research. We propose Benchopt, a collaborative framework to automatize, publish and reproduce optimization benchmarks in machine learning across programming languages and hardware architectures. Benchopt simplifies benchmarking for the community by providing an off-the-shelf tool for running, sharing and extending experiments. To demonstrate its broad usability, we showcase benchmarks on three standard ML tasks: $\ell_2$-regularized logistic regression, Lasso and ResNet18 training for image classification. These benchmarks highlight key practical findings that give a more nuanced view of state-of-the-art for these problems, showing that for practical evaluation, the devil is in the details.

PDF Details

NeurIPS Conference 2022 Conference Paper

Do Residual Neural Networks discretize Neural Ordinary Differential Equations?

Michael Sander
Pierre Ablin
Gabriel Peyré

Neural Ordinary Differential Equations (Neural ODEs) are the continuous analog of Residual Neural Networks (ResNets). We investigate whether the discrete dynamics defined by a ResNet are close to the continuous one of a Neural ODE. We first quantify the distance between the ResNet's hidden state trajectory and the solution of its corresponding Neural ODE. Our bound is tight and, on the negative side, does not go to $0$ with depth $N$ if the residual functions are not smooth with depth. On the positive side, we show that this smoothness is preserved by gradient descent for a ResNet with linear residual functions and small enough initial loss. It ensures an implicit regularization towards a limit Neural ODE at rate $\frac1N$, uniformly with depth and optimization time. As a byproduct of our analysis, we consider the use of a memory-free discrete adjoint method to train a ResNet by recovering the activations on the fly through a backward pass of the network, and show that this method theoretically succeeds at large depth if the residual functions are Lipschitz with the input. We then show that Heun's method, a second order ODE integration scheme, allows for better gradient estimation with the adjoint method when the residual functions are smooth with depth. We experimentally validate that our adjoint method succeeds at large depth, and that Heun’s method needs fewer layers to succeed. We finally use the adjoint method successfully for fine-tuning very deep ResNets without memory consumption in the residual layers.

PDF Details

ICML Conference 2021 Conference Paper

Kernel Stein Discrepancy Descent

Anna Korba
Pierre-Cyril Aubin-Frankowski
Szymon Majewski
Pierre Ablin

Among dissimilarities between probability distributions, the Kernel Stein Discrepancy (KSD) has received much interest recently. We investigate the properties of its Wasserstein gradient flow to approximate a target probability distribution $\pi$ on $\mathbb{R}^d$, known up to a normalization constant. This leads to a straightforwardly implementable, deterministic score-based method to sample from $\pi$, named KSD Descent, which uses a set of particles to approximate $\pi$. Remarkably, owing to a tractable loss function, KSD Descent can leverage robust parameter-free optimization schemes such as L-BFGS; this contrasts with other popular particle-based schemes such as the Stein Variational Gradient Descent algorithm. We study the convergence properties of KSD Descent and demonstrate its practical relevance. However, we also highlight failure cases by showing that the algorithm can get stuck in spurious local minima.

Details

ICML Conference 2021 Conference Paper

Momentum Residual Neural Networks

Michael Eli Sander
Pierre Ablin
Mathieu Blondel
Gabriel Peyré

The training of deep residual neural networks (ResNets) with backpropagation has a memory cost that increases linearly with respect to the depth of the network. A simple way to circumvent this issue is to use reversible architectures. In this paper, we propose to change the forward rule of a ResNet by adding a momentum term. The resulting networks, momentum residual neural networks (MomentumNets), are invertible. Unlike previous invertible architectures, they can be used as a drop-in replacement for any existing ResNet block. We show that MomentumNets can be interpreted in the infinitesimal step size regime as second-order ordinary differential equations (ODEs) and exactly characterize how adding momentum progressively increases the representation capabilities of MomentumNets: they can learn any linear mapping up to a multiplicative factor, while ResNets cannot. In a learning to optimize setting, where convergence to a fixed point is required, we show theoretically and empirically that our method succeeds while existing invertible architectures fail. We show on CIFAR and ImageNet that MomentumNets have the same accuracy as ResNets, while having a much smaller memory footprint, and show that pre-trained MomentumNets are promising for fine-tuning models.

Details

JMLR Journal 2021 Journal Article

mvlearn: Multiview Machine Learning in Python

Ronan Perry
Gavin Mischler
Richard Guo
Theodore Lee
Alexander Chang
Arman Koul
Cameron Franz
Hugo Richard

As data are generated more and more from multiple disparate sources, multiview data sets, where each sample has features in distinct views, have grown in recent years. However, no comprehensive package exists that enables non-specialists to use these methods easily. mvlearn is a Python library which implements the leading multiview machine learning methods. Its simple API closely follows that of scikit-learn for increased ease-of-use. The package can be installed from Python Package Index (PyPI) and the conda package manager and is released under the MIT open-source license. The documentation, detailed examples, and all releases are available at https://mvlearn.github.io/. [abs] [ pdf ][ bib ] [ code ] &copy JMLR 2021. ( edit, beta )

PDF Details

NeurIPS Conference 2021 Conference Paper

Shared Independent Component Analysis for Multi-Subject Neuroimaging

Hugo Richard
Pierre Ablin
Bertrand Thirion
Alexandre Gramfort
Aapo Hyvarinen

We consider shared response modeling, a multi-view learning problem where one wants to identify common components from multiple datasets or views. We introduce Shared Independent Component Analysis (ShICA) that models eachview as a linear transform of shared independent components contaminated by additive Gaussian noise. We show that this model is identifiable if the components are either non-Gaussian or have enough diversity in noise variances. We then show that in some cases multi-set canonical correlation analysis can recover the correct unmixing matrices, but that even a small amount of sampling noise makes Multiset CCA fail. To solve this problem, we propose to use joint diagonalization after Multiset CCA, leading to a new approach called ShICA-J. We show via simulations that ShICA-J leads to improved results while being very fast to fit. While ShICA-J is based on second-order statistics, we further propose to leverage non-Gaussianity of the components using a maximum-likelihood method, ShICA-ML, that is both more accurate and more costly. Further, ShICA comes with a principled method for shared components estimation. Finally, we provide empirical evidence on fMRI and MEG datasets that ShICA yields more accurate estimation of the componentsthan alternatives.

PDF Details

NeurIPS Conference 2020 Conference Paper

Modeling Shared responses in Neuroimaging Studies through MultiView ICA

Hugo Richard
Luigi Gresele
Aapo Hyvarinen
Bertrand Thirion
Alexandre Gramfort
Pierre Ablin

Group studies involving large cohorts of subjects are important to draw general conclusions about brain functional organization. However, the aggregation of data coming from multiple subjects is challenging, since it requires accounting for large variability in anatomy, functional topography and stimulus response across individuals. Data modeling is especially hard for ecologically relevant conditions such as movie watching, where the experimental setup does not imply well-defined cognitive operations. We propose a novel MultiView Independent Component Analysis (ICA) model for group studies, where data from each subject are modeled as a linear combination of shared independent sources plus noise. Contrary to most group-ICA procedures, the likelihood of the model is available in closed form. We develop an alternate quasi-Newton method for maximizing the likelihood, which is robust and converges quickly. We demonstrate the usefulness of our approach first on fMRI data, where our model demonstrates improved sensitivity in identifying common sources among subjects. Moreover, the sources recovered by our model exhibit lower between-sessions variability than other methods. On magnetoencephalography (MEG) data, our method yields more accurate source localization on phantom data. Applied on 200 subjects from the Cam-CAN dataset, it reveals a clear sequence of evoked activity in sensor and source space.

PDF Details

YNIMG Journal 2020 Journal Article

Predictive regression modeling with MEG/EEG: from source power to signals and cognitive states

David Sabbagh
Pierre Ablin
Gaël Varoquaux
Alexandre Gramfort
Denis A. Engemann

Predicting biomedical outcomes from Magnetoencephalography and Electroencephalography (M/EEG) is central to applications like decoding, brain-computer-interfaces (BCI) or biomarker development and is facilitated by supervised machine learning. Yet, most of the literature is concerned with classification of outcomes defined at the event-level. Here, we focus on predicting continuous outcomes from M/EEG signal defined at the subject-level, and analyze about 600 MEG recordings from Cam-CAN dataset and about 1000 EEG recordings from TUH dataset. Considering different generative mechanisms for M/EEG signals and the biomedical outcome, we propose statistically-consistent predictive models that avoid source-reconstruction based on the covariance as representation. Our mathematical analysis and ground-truth simulations demonstrated that consistent function approximation can be obtained with supervised spatial filtering or by embedding with Riemannian geometry. Additional simulations revealed that Riemannian methods were more robust to model violations, in particular geometric distortions induced by individual anatomy. To estimate the relative contribution of brain dynamics and anatomy to prediction performance, we propose a novel model inspection procedure based on biophysical forward modeling. Applied to prediction of outcomes at the subject-level, the analysis revealed that the Riemannian model better exploited anatomical information while sensitivity to brain dynamics was similar across methods. We then probed the robustness of the models across different data cleaning options. Environmental denoising was globally important but Riemannian models were strikingly robust and continued performing well even without preprocessing. Our results suggest each method has its niche: supervised spatial filtering is practical for event-level prediction while the Riemannian model may enable simple end-to-end learning.

Details DOI

ICML Conference 2020 Conference Paper

Super-efficiency of automatic differentiation for functions defined as a minimum

Pierre Ablin
Gabriel Peyré
Thomas Moreau 0001

In min-min optimization or max-min optimization, one has to compute the gradient of a function defined as a minimum. In most cases, the minimum has no closed-form, and an approximation is obtained via an iterative algorithm. There are two usual ways of estimating the gradient of the function: using either an analytic formula obtained by assuming exactness of the approximation, or automatic differentiation through the algorithm. In this paper, we study the asymptotic error made by these estimators as a function of the optimization error. We find that the error of the automatic estimator is close to the square of the error of the analytic estimator, reflecting a super-efficiency phenomenon. The convergence of the automatic estimator greatly depends on the convergence of the Jacobian of the algorithm. We analyze it for gradient descent and stochastic gradient descent and derive convergence rates for the estimators in these cases. Our analysis is backed by numerical experiments on toy problems and on Wasserstein barycenter computation. Finally, we discuss the computational complexity of these estimators and give practical guidelines to chose between them.

Details

NeurIPS Conference 2019 Conference Paper

Learning step sizes for unfolded sparse coding

Pierre Ablin
Thomas Moreau
Mathurin Massias
Alexandre Gramfort

Sparse coding is typically solved by iterative optimization techniques, such as the Iterative Shrinkage-Thresholding Algorithm (ISTA). Unfolding and learning weights of ISTA using neural networks is a practical way to accelerate estimation. In this paper, we study the selection of adapted step sizes for ISTA. We show that a simple step size strategy can improve the convergence rate of ISTA by leveraging the sparsity of the iterates. However, it is impractical in most large-scale applications. Therefore, we propose a network architecture where only the step sizes of ISTA are learned. We demonstrate that for a large class of unfolded algorithms, if the algorithm converges to the solution of the Lasso, its last layers correspond to ISTA with learned step sizes. Experiments show that our method is competitive with state-of-the-art networks when the solutions are sparse enough.

PDF Details

NeurIPS Conference 2019 Conference Paper

Manifold-regression to predict from MEG/EEG brain signals without source modeling

David Sabbagh
Pierre Ablin
Gael Varoquaux
Alexandre Gramfort
Denis Engemann

Magnetoencephalography and electroencephalography (M/EEG) can reveal neuronal dynamics non-invasively in real-time and are therefore appreciated methods in medicine and neuroscience. Recent advances in modeling brain-behavior relationships have highlighted the effectiveness of Riemannian geometry for summarizing the spatially correlated time-series from M/EEG in terms of their covariance. However, after artefact-suppression, M/EEG data is often rank deficient which limits the application of Riemannian concepts. In this article, we focus on the task of regression with rank-reduced covariance matrices. We study two Riemannian approaches that vectorize the M/EEG covariance between sensors through projection into a tangent space. The Wasserstein distance readily applies to rank-reduced data but lacks affine-invariance. This can be overcome by finding a common subspace in which the covariance matrices are full rank, enabling the affine-invariant geometric distance. We investigated the implications of these two approaches in synthetic generative models, which allowed us to control estimation bias of a linear model for prediction. We show that Wasserstein and geometric distances allow perfect out-of-sample prediction on the generative models. We then evaluated the methods on real data with regard to their effectiveness in predicting age from M/EEG covariance matrices. The findings suggest that the data-driven Riemannian methods outperform different sensor-space estimators and that they get close to the performance of biophysics-driven source-localization model that requires MRI acquisitions and tedious data processing. Our study suggests that the proposed Riemannian methods can serve as fundamental building-blocks for automated large-scale analysis of M/EEG.

PDF Details

JBHI Journal 2018 Journal Article

Statistical Shape Modeling of the Left Ventricle: Myocardial Infarct Classification Challenge

Avan Suinesiaputra
Jan Dhooge
Nicolas Duchateau
Jan Ehrhardt
Alejandro F. Frangi
Ali Gooya
Vicente Grau
Karim Lekadir

Statistical shape modeling is a powerful tool for visualizing and quantifying geometric and functional patterns of the heart. After myocardial infarction (MI), the left ventricle typically remodels in response to physiological challenges. Several methods have been proposed in the literature to describe statistical shape changes. Which method best characterizes the left ventricular remodeling after MI is an open research question. A better descriptor of remodeling is expected to provide a more accurate evaluation of disease status in MI patients. We therefore designed a challenge to test shape characterization in MI given a set of three-dimensional left ventricular surface points. The training set comprised 100 MI patients, and 100 asymptomatic volunteers (AV). The challenge was initiated in 2015 at the Statistical Atlases and Computational Models of the Heart workshop, in conjunction with the MICCAI conference. The training set with labels was provided to participants, who were asked to submit the likelihood of MI from a different (validation) set of 200 cases (100 AV and 100 MI). Sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve were used as the outcome measures. The goals of this challenge were to 1) establish a common dataset for evaluating statistical shape modeling algorithms in MI, and 2) test whether statistical shape modeling provides additional information characterizing MI patients over standard clinical measures. Eleven groups with a wide variety of classification and feature extraction approaches participated in this challenge. All methods achieved excellent classification results with accuracy ranges from 0. 83 to 0. 98. The areas under the receiver operating characteristic curves were all above 0. 90. Four methods showed significantly higher performance than standard clinical measures. The dataset and software for evaluation are available from the Cardiac Atlas Project website. 1 1 http://www.cardiacatlas.org.

Details DOI