Search

International audience; Importance sampling (IS) is a powerful Monte Carlo methodology for the approximation of intractable integrals, very often involving a target probability distribution. The performance of IS heavily depends on the appropriate selection of the proposal distributions where the samples are simulated from. In this paper, we propose an adaptive importance sampler, called GRAMIS, that iteratively improves the set of proposals. The algorithm exploits geometric information of the target to adapt the location and scale parameters of those proposals. Moreover, in order to allow for a cooperative adaptation, a repulsion term is introduced that favors a coordinated exploration of the state space. This translates into a more diverse exploration and a better approximation of the target via the mixture of proposals. Moreover, we provide a theoretical justification of the repulsion term. We show the good performance of GRAMIS in two problems where the target has a challenging shape and cannot be easily approximated by a standard uni-modal proposal.

Denoising, detrending, deconvolution: usual restoration tasks, traditionally decoupled. Coupled formulations entail complex ill-posed inverse problems. We propose PENDANTSS for joint trend removal and blind deconvolution of sparse peak-like signals. It blends a parsimonious prior with the hypothesis that smooth trend and noise can somewhat be separated by low-pass filtering. We combine the generalized pseudo-norm ratio SOOT/SPOQ sparse penalties $\ell_p/\ell_q$ with the BEADS ternary assisted source separation algorithm. This results in a both convergent and efficient tool, with a novel Trust-Region block alternating variable metric forward-backward approach. It outperforms comparable methods, when applied to typically peaked analytical chemistry signals. Reproducible code is provided: https://github.com/paulzhengfr/PENDANTSS., Comment: in French language

Co-administration of two or more drugs simultaneously can result in adverse drug reactions. Identifying drug-drug interactions (DDIs) is necessary, especially for drug development and for repurposing old drugs. DDI prediction can be viewed as a matrix completion task, for which matrix factorization (MF) appears as a suitable solution. This paper presents a novel Graph Regularized Probabilistic Matrix Factorization (GRPMF) method, which incorporates expert knowledge through a novel graph-based regularization strategy within an MF framework. An efficient and sounded optimization algorithm is proposed to solve the resulting non-convex problem in an alternating fashion. The performance of the proposed method is evaluated through the DrugBank dataset, and comparisons are provided against state-of-the-art techniques. The results demonstrate the superior performance of GRPMF when compared to its counterparts.

International audience

International audience; Denoising, detrending, deconvolution: usual restoration tasks, traditionally decoupled. Coupled formulations entail complex ill-posed inverse problems. We propose PENDANTSS for joint trend removal and blind deconvolution of sparse peak-like signals. It blends a parsimonious prior with the hypothesis that smooth trend and noise can somewhat be separated by low-pass filtering. We combine the generalized quasi-norm ratio SOOT/SPOQ sparse penalties $\ell_p/\ell_q$ with the BEADS ternary assisted source separation algorithm. This results in a both convergent and efficient tool, with a novel Trust-Region block alternating variable metric forward-backward approach. It outperforms comparable methods, when applied to typically peaked analytical chemistry signals. Reproducible code is provided.

International audience; Lifting-based wavelet transform has been extensively used for efficient compression of various types of visual data. Generally, the performance of such coding schemes strongly depends on the lifting operators used, namely the prediction and update filters. Unlike conventional schemes based on linear filters, we propose, in this paper, to learn these operators by exploiting neural networks. More precisely, a classical Fully Connected Neural Network (FCNN) architecture is firstly employed to perform the prediction and update. Then, we propose to improve this FCNN-based Lifting Scheme (LS) in order to better take into account the input image to be encoded. Thus, a novel dynamical FCNN model is developed, making the learning process adaptive to the input image contents for which two adaptive learning techniques are proposed. While the first one resorts to an iterative algorithm where the computation of two kinds of variables is performed in an alternating manner, the second learning method aims to learn the model parameters directly through a reformulation of the loss function. Experimental results carried out on various test images show the benefits of the proposed approaches in the context of lossy and lossless image compression.

Stochastic differentiable approximation schemes are widely used for solving high dimensional problems. Most of existing methods satisfy some desirable properties, including conditional descent inequalities, and almost sure (a.s.) convergence guarantees on the objective function, or on the involved gradient. However, for non-convex objective functions, a.s. convergence of the iterates, i.e., the stochastic process, to a critical point is usually not guaranteed, and remains an important challenge. In this article, we develop a framework to bridge the gap between descent-type inequalities and a.s. convergence of the associated stochastic process. Leveraging a novel Kurdyka-Łojasiewicz property, we show convergence guarantees of stochastic processes under mild assumptions on the objective function. We also provide examples of stochastic algorithms benefiting from the proposed framework and derive a.s. convergence guarantees on the iterates.

Lipschitz bounded neural networks are certifiably robust and have a good trade-off between clean and certified accuracy. Existing Lipschitz bounding methods train from scratch and are limited to moderately sized networks (< 6M parameters). They require a fair amount of hyper-parameter tuning and are computationally prohibitive for large networks like Vision Transformers (5M to 660M parameters). Obtaining certified robustness of transformers is not feasible due to the non-scalability and inflexibility of the current methods. This work presents CertViT, a two-step proximal-projection method to achieve certified robustness from pre-trained weights. The proximal step tries to lower the Lipschitz bound and the projection step tries to maintain the clean accuracy of pre-trained weights. We show that CertViT networks have better certified accuracy than state-of-the-art Lipschitz trained networks. We apply CertViT on several variants of pre-trained vision transformers and show adversarial robustness using standard attacks. Code : https://github.com/sagarverma/transformer-lipschitz, Preprint work. 13 pages, 3 figures, https://github.com/sagarverma/transformer-lipschitz

Granger causality (GC) is often considered not an actual form of causality. Still, it is arguably the most widely used method to assess the predictability of a time series from another one. Granger causality has been widely used in many applied disciplines, from neuroscience and econometrics to Earth sciences. We revisit GC under a graphical perspective of state-space models. For that, we use GraphEM, a recently presented expectation-maximisation algorithm for estimating the linear matrix operator in the state equation of a linear-Gaussian state-space model. Lasso regularisation is included in the M-step, which is solved using a proximal splitting Douglas-Rachford algorithm. Experiments in toy examples and challenging climate problems illustrate the benefits of the proposed model and inference technique over standard Granger causality methods., Comment: 4 pages, 2 figures, 3 tables, CausalStats23: When Causal Inference meets Statistical Analysis, April 17-21, 2023, Paris, France

We consider the minimization of a differentiable Lipschitz gradient but non necessarily convex, function F defined on R N. We propose an accelerated gradient descent approach which combines three strategies, namely (i) a variable metric derived from the majorization-minimization principle ; (ii) a subspace strategy incorporating information from the past iterates ; (iii) a block alternating update. Under the assumption that F satisfies the Kurdyka-Łojasiewicz property, we give conditions under which the sequence generated by the resulting block majorize-minimize subspace algorithm converges to a critical point of the objective function, and we exhibit convergence rates for its iterates.

International audience; Stochastic gradient optimization methods are broadly used to minimize non-convex smooth objective functions, for instance when training deep neural networks. However, theoretical guarantees on the asymptotic behaviour of these methods remain scarce. Especially, ensuring almost-sure convergence of the iterates to a stationary point is quite challenging. In this work, we introduce a new Kurdyka-Łojasiewicz theoretical framework to analyze asymptotic behavior of stochastic gradient descent (SGD) schemes when minimizing non-convex smooth objectives. In particular, our framework provides new almost-sure convergence results, on iterates generated by any SGD method satisfying mild conditional descent conditions. We illustrate the proposed framework by means of several toy simulation examples. We illustrate the role of the considered theoretical assumptions, and investigate how SGD iterates are impacted whether these assumptions are either fully or partially satisfied.

International audience; In this paper, a neural network approach is introduced to estimate non-noisy speed and torque from noisy measured currents and voltages in induction motors with Variable Speed Drives. The proposed estimation method is comprised of a neural speed-torque estimator and a neural signal denoiser. A new training strategy is introduced that combines large amount of simulated data and a small amount of real world data. The proposed denoiser does not require non-noisy ground truth data for training, and instead uses classification labels which are easily generated from real-world data. This approach improves upon existing noise removal techniques by learning to denoise as well as classify noisy signals into static and dynamic parts. The proposed neural network based denoiser generates clean estimates of currents and voltages which are then used as inputs to the neural network estimator of speed and torque. Extensive experiments show that the proposed joint denoising-estimation strategy performs very well on real data benchmarks. The proposed denoising method is shown to outperform several widely used denoising methods and a proper ablation study of the proposed method is conducted.

We investigate inexact proximity operators for weakly convex functions. To this aim, we derive sum rules for proximal ε-subdifferentials, by incorporating the moduli of weak convexity of the functions into the respective formulas. This allows us to investigate inexact proximity operators for weakly convex functions in terms of proximal ε-subdifferentials.

We leverage path differentiability and a recent result on nonsmooth implicit differentiation calculus to give sufficient conditions ensuring that the solution to a monotone inclusion problem will be path differentiable, with formulas for computing its generalized gradient. A direct consequence of our result is that these solutions happen to be differentiable almost everywhere. Our approach is fully compatible with automatic differentiation and comes with assumptions which are easy to check, roughly speaking: semialgebraicity and strong monotonicity. We illustrate the scope of our results by considering three fundamental composite problem settings: strongly convex problems, dual solutions to convex minimization problems and primal-dual solutions to min-max problems.

In this paper, we introduce a variational Bayesian algorithm (VBA) for image blind deconvolution. Our generic framework incorporates smoothness priors on the unknown blur/image and possible affine constraints (e.g., sum to one) on the blur kernel. One of our main contributions is the integration of VBA within a neural network paradigm, following an unrolling methodology. The proposed architecture is trained in a supervised fashion, which allows us to optimally set two key hyperparameters of the VBA model and lead to further improvements in terms of resulting visual quality. Various experiments involving grayscale/color images and diverse kernel shapes, are performed. The numerical examples illustrate the high performance of our approach when compared to state-of-the-art techniques based on optimization, Bayesian estimation, or deep learning., Comment: 13 pages

International audience; Standard few-shot benchmarks are often built upon simplifying assumptions on the query sets, which may not always hold in practice. In particular, for each task at testing time, the classes effectively present in the unlabeled query set are known a priori, and correspond exactly to the set of classes represented in the labeled support set. We relax these assumptions and extend current benchmarks, so that the query-set classes of a given task are unknown, but just belong to a much larger set of possible classes. Our setting could be viewed as an instance of the challenging yet practical problem of extremely imbalanced K-way classification, K being much larger than the values typically used in standard benchmarks, and with potentially irrelevant supervision from the support set. Expectedly, our setting incurs drops in the performances of state-of-the-art methods. Motivated by these observations, we introduce a PrimAl Dual Minimum Description LEngth (PADDLE) formulation, which balances data-fitting accuracy and model complexity for a given few-shot task, under supervision constraints from the support set. Our constrained MDL-like objective promotes competition among a large set of possible classes, preserving only effective classes that befit better the data of a few-shot task. It is hyperparameter free, and could be applied on top of any base-class training. Furthermore, we derive a fast block coordinate descent algorithm for optimizing our objective, with convergence guarantee, and a linear computational complexity at each iteration. Comprehensive experiments over the standard few-shot datasets and the more realistic and challenging i-Nat dataset show highly competitive performances of our method, more so when the numbers of possible classes in the tasks increase. Our code is publicly available at https://github.com/SegoleneMartin/PADDLE.

Graph Representation Learning (GRL) has become a key paradigm in network analysis, with a plethora of interdisciplinary applications. As the scale of networks increases, most of the widely used learning-based graph representation models also face computational challenges. While there is a recent effort toward designing algorithms that solely deal with scalability issues, most of them behave poorly in terms of accuracy on downstream tasks. In this paper, we aim to study models that balance the trade-off between efficiency and accuracy. In particular, we propose NodeSig, a scalable model that computes binary node representations. NodeSig exploits random walk diffusion probabilities via stable random projections towards efficiently computing embeddings in the Hamming space. Our extensive experimental evaluation on various networks has demonstrated that the proposed model achieves a good balance between accuracy and efficiency compared to well-known baseline models on the node classification and link prediction tasks., Comment: The manuscript was accepted to the 2022 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)

Publisher Copyright: © 2022 The Author(s) This paper presents a multimodal image fusion method using a novel decomposition model based on coupled dictionary learning. The proposed method is general and can be used for a variety of imaging modalities. In particular, the images to be fused are decomposed into correlated and uncorrelated components using sparse representations with identical supports and a Pearson correlation constraint, respectively. The resulting optimization problem is solved by an alternating minimization algorithm. Contrary to other learning-based fusion methods, the proposed approach does not require any training data, and the correlated features are extracted online from the data itself. By preserving the uncorrelated components in the fused images, the proposed fusion method significantly improves on current fusion approaches in terms of maintaining the texture details and modality-specific information. The maximum-absolute-value rule is used for the fusion of correlated components only. This leads to an enhanced contrast-resolution without causing intensity attenuation or loss of important information. Experimental results show that the proposed method achieves superior performance in terms of both visual and objective evaluations compared to state-of-the-art image fusion methods.

BackgroundAnti-PD-(L)1 treatment is indicated for patients with mismatch repair-deficient (MMRD) tumors, regardless of tumor origin. However, the response rate is highly heterogeneous across MMRD tumors. The objective of the study is to find a score that predicts anti-PD-(L)1 response in patients with MMRD tumors.MethodsSixty-one patients with various origin of MMRD tumors and treated with anti-PD-(L)1 were retrospectively included in this study. An expert radiologist annotated all tumors present at the baseline and first evaluation CT-scans for all the patients by circumscribing them on their largest axial axis (single slice), allowing us to compute an approximation of their tumor volume. In total, 2120 lesions were annotated, which led to the computation of the total tumor volume for each patient. The RECIST sum of target lesions’ diameters and neutrophile-to-lymphocyte (NLR) were also reported at both examinations. These parameters were determined at baseline and first evaluation and the variation between the first evaluation and baseline was calculated, to determine a comprehensive score for overall survival (OS) and progression-free survival (PFS).ResultsTotal tumor volume at baseline was found to be significantly correlated to the OS (p-value: 0.005) and to the PFS (p-value:ConclusionTotal tumor volume appears to be a prognostic biomarker of anti-PD-(L)1 response to immunotherapy in metastatic patients with MMRD tumors. Combining tumor volume and NLR with a simple concordance score stratifies patients well according to their survival and offers a good predictive measure of response to immunotherapy.

BackgroundGene expression is regulated at different molecular levels, including chromatin accessibility, transcription, RNA maturation, and transport. These regulatory mechanisms have strong connections with cellular metabolism. In order to study the cellular system and its functioning, omics data at each molecular level can be generated and efficiently integrated. Here, we proposeBRANEnet, a novel multi-omics integration framework for multilayer heterogeneous networks.BRANEnetis an expressive, scalable, and versatile method to learn node embeddings, leveraging random walk information within a matrix factorization framework. Our goal is to efficiently integrate multi-omics data to study different regulatory aspects of multilayered processes that occur in organisms. We evaluate our framework using multi-omics data ofSaccharomyces cerevisiae, a well-studied yeast model organism.ResultsWe testBRANEneton transcriptomics (RNA-seq) and targeted metabolomics (NMR) data for wild-type yeast strain during a heat-shock time course of 0, 20, and 120 min. Our framework learns features for differentially expressed bio-molecules showing heat stress response. We demonstrate the applicability of the learned features for targeted omics inference tasks: transcription factor (TF)-target prediction, integrated omics network (ION) inference, and module identification. The performance ofBRANEnetis compared to existing network integration methods. Our model outperforms baseline methods by achieving high prediction scores for a variety of downstream tasks.

In the last ten years, Convolutional Neural Networks (CNNs) have formed the basis of deep-learning architectures for most computer vision tasks. However, they are not necessarily optimal. For example, mathematical morphology is known to be better suited to deal with binary images. In this work, we create a morphological neural network that handles binary inputs and outputs. We propose their construction inspired by CNNs to formulate layers adapted to such images by replacing convolutions with erosions and dilations. We give explainable theoretical results on whether or not the resulting learned networks are indeed morphological operators. We present promising experimental results designed to learn basic binary operators, and we have made our code publicly available online., Comment: Preprint of a submission to ICIP 2022. 7 pages. 4 figures

International audience; For best medical imaging application results, learning-based approaches such as deep learning necessitate specific, extensive and precise annotations. Outside well-curated public benchmarks, these are rarely available in practice, and so it becomes necessary to use less-than-perfect annotations. One way of compensating for this is the embedding of anatomical knowledge. Complementing this, there is the incremental semi-supervised learning technique, whereby a small amount of annotations can be used to derive more and superior labels.In this article, we illustrate this approach on a deep learning system to help radiologists and rheumatologists finely and interactively assess MRI scans of the sacro-iliac joint in order to correctly diagnose Axial Spondyloarthritis. Our model is trained initially on a relatively small set of images with promising results, on par with expert opinion and generalizable to new datasets.

Artificial intelligence (AI) is increasingly being studied in musculoskeletal oncology imaging. AI has been applied to both primary and secondary bone tumors and assessed for various predictive tasks that include detection, segmentation, classification, and prognosis. Still, in the field of clinical research, further efforts are needed to improve AI reproducibility and reach an acceptable level of evidence in musculoskeletal oncology. This review describes the basic principles of the most common AI techniques, including machine learning, deep learning and radiomics. Then, recent developments and current results of AI in the field of musculoskeletal oncology are presented. Finally, limitations and future perspectives of AI in this field are discussed.

Histopathology whole slide images (WSIs) play a very important role in clinical studies and serve as the gold standard for many cancer diagnoses. However, generating automatic tools for processing WSIs is challenging due to their enormous sizes. Currently, to deal with this issue, conventional methods rely on a multiple instance learning (MIL) strategy to process a WSI at patch level. Although effective, such methods are computationally expensive, because tiling a WSI into patches takes time and does not explore the spatial relations between these tiles. To tackle these limitations, we propose a locally supervised learning framework which processes the entire slide by exploring the entire local and global information that it contains. This framework divides a pre-trained network into several modules and optimizes each module locally using an auxiliary model. We also introduce a random feature reconstruction unit (RFR) to preserve distinguishing features during training and improve the performance of our method by 1% to 3%. Extensive experiments on three publicly available WSI datasets: TCGA-NSCLC, TCGA-RCC and LKS, highlight the superiority of our method on different classification tasks. Our method outperforms the state-of-the-art MIL methods by 2% to 5% in accuracy, while being 7 to 10 times faster. Additionally, when dividing it into eight modules, our method requires as little as 20% of the total gpu memory required by end-to-end training. Our code is available at https://github.com/cvlab-stonybrook/local_learning_wsi., Accepted to MICCAI 2022 Oral

International audience

International audience; State-space models (SSM) are central to describe time-varying complex systems in countless signal processing applications such as remote sensing, networks, biomedicine, and finance to name a few. Inference and prediction in SSMs are possible when the model parameters are known, which is rarely the case. The estimation of these parameters is crucial, not only for performing statistical analysis, but also for uncovering the underlying structure of complex phenomena. In this paper, we focus on the linear-Gaussian model, arguably the most celebrated SSM, and particularly in the challenging task of estimating the transition matrix that encodes the Markovian dependencies in the evolution of the multi-variate state. We introduce a novel perspective by relating this matrix to the adjacency matrix of a directed graph, also interpreted as the causal relationship among state dimensions in the Granger-causality sense. Under this perspective, we propose a new method called GraphEM based on the well sounded expectation-maximization (EM) methodology for inferring the transition matrix jointly with the smoothing/filtering of the observed data. We propose an advanced convex optimization solver relying on a consensus-based implementation of a proximal splitting strategy for solving the M-step. This approach enables an efficient and versatile processing of various sophisticated priors on the graph structure, such as parsimony constraints, while benefiting from convergence guarantees. We demonstrate the good performance and the interpretable results of GraphEM by means of two sets of numerical examples.

International audience; Both theoretical and practical problems in deep learning classification require solutions for assessing uncertainty prediction but current state-of-the-art methods in this area are computationally expensive. In this paper, we propose a new confidence measure dubbed Over-MAP that utilizes a measure of overlap between structural attention mechanisms and segmentation methods, that is of particular interest in accurate fine-grained contexts. We show that this classification confidence increases with the degree of overlap. The associated confidence and identification tools are conceptually simple, efficient, and of high practical interest as they allow for weeding out misleading examples in training data. Our measure is currently deployed in the real-world on widely used platforms to annotate large-scale data efficiently.

International audience; Optimization problems arising in signal and image processing involve an increasingly large number of variables. In addition to the curse of dimensionality, another difficulty to overcome is that the cost function usually reads as the sum of several loss/regularization terms, which are non-necessarily smooth and possibly composed with large-size linear operators. Proximal splitting approaches are fundamental tools to address such problems, with demonstrated efficiency in many applicative fields. In this paper, we present a new distributed algorithm for computing the proximity operator of a sum of non-necessarily smooth convex functions composed with arbitrary linear operators. Our algorithm relies on a primal-dual splitting strategy, and benefits from established convergence guaranties. Each involved function is associated with a node of a hypergraph, with the ability to communicate with neighboring nodes sharing the same hyperedge. Thanks to this structure, our method can be efficiently implemented on modern parallel computing architectures, distributing the computations on multiple nodes or machines, with controlled requirements for synchronization steps. Good numerical performance and scalability properties are demonstrated on a problem of video sequence denoising. Our code implemented in Julia is made available at \url{https://github.com/MarinENSTA/distributed_julia_denoising}.

International audience; ObjectivesThe aim of this study is to evaluate a deep learning method designed to increase the contrast-to-noise ratio in contrast-enhanced gradient echo T1-weighted brain MRI acquisitions. The processed images are quantitatively evaluated in terms of lesion detection performance.Materials and methodsA total of 250 multiparametric brain MRIs, acquired between November 2019 and March 2021 at Gustave Roussy Cancer Campus (Villejuif, France), were considered for inclusion in this retrospective monocentric study. Independent training (107 cases, age 55y±14, 58 women) and test (79 cases, age 59y±14, 41 women) samples were defined. Patients had glioma, brain metastasis, meningioma, or no enhancing lesion. Gradient echo and turbo spin echo with variable flip angles postcontrast T1 sequences were acquired in all cases. For the cases that formed the training sample, “low-dose” postcontrast gradient echo T1 images using 0.025mmol/kg injections of contrast agent were also acquired. A deep neural network was trained to synthetically enhance the low-dose T1 acquisitions, taking standard-dose T1 MRI as reference. Once trained, the contrast enhancement network was used to process the test gradient echo T1 images. A read was then performed by two experienced neuroradiologists to evaluate the original and processed T1 MRI sequences in terms of contrast enhancement and lesion detection performance, taking the turbo spin echo sequences as reference.ResultsThe processed images were superior to the original gradient echo and reference turbo spin echo T1 sequences in terms of contrast-to-noise ratio (44.5 versus 9.1 and 16.8, p.99). The same effect was observed when considering all lesions larger than 5mm: sensitivity increased from 70% to 85% (p

International audience

International audience; The identification of parameters of spatially variant blurs given a clean image and its blurry noisy version is a challenging inverse problem of interest in many application fields, such as biological microscopy and astronomical imaging. In this paper, we consider a parametric model of the blur and introduce an 1D state-space model to describe the statistical dependence among the neighboring kernels. We apply a Bayesian approach to estimate the posterior distribution of the kernel parameters given the available data. Since this posterior is intractable for most realistic models, we propose to approximate it through a sequential Monte Carlo approach by processing all data in a sequential and efficient manner. Additionally, we propose a new sampling method to alleviate the particle degeneracy problem, which is present in approximate Bayesian filtering, particularly in challenging concentrated posterior distributions. The considered method allows us to process sequentially image patches at a reasonable computational and memory costs. Moreover, the probabilistic approach we adopt in this paper provides uncertainty quantification which is useful for image restoration. The practical experimental results illustrate the improved estimation performance of our novel approach, demonstrating also the benefits of exploiting the spatial structure the parametric blurs in the considered models.

The SARS-COV-2 pandemic has put pressure on intensive care units, so that identifying predictors of disease severity is a priority. We collect 58 clinical and biological variables, and chest CT scan data, from 1003 coronavirus-infected patients from two French hospitals. We train a deep learning model based on CT scans to predict severity. We then construct the multimodal AI-severity score that includes 5 clinical and biological variables (age, sex, oxygenation, urea, platelet) in addition to the deep learning model. We show that neural network analysis of CT-scans brings unique prognosis information, although it is correlated with other markers of severity (oxygenation, LDH, and CRP) explaining the measurable but limited 0.03 increase of AUC obtained when adding CT-scan information to clinical variables. Here, we show that when comparing AI-severity with 11 existing severity scores, we find significantly improved prognosis performance; AI-severity can therefore rapidly become a reference scoring approach., The SARS-COV-2 pandemic has put pressure on intensive care units, so that predicting severe deterioration early is a priority. Here, the authors develop a multimodal severity score including clinical and imaging features that has significantly improved prognostic performance in two validation datasets compared to previous scores.

International audience; In this work we address the problem of short-term load forecasting. We propose a generalization of the linear state-space model where the evolution of the state and the observation matrices is unknown. The proposed blind Kalman filter algorithm proceeds via alternating the estimation of these unknown matrices and the inference of the state, within the framework of expectation-maximization. A mini-batch processing strategy is introduced to allow on-the-fly forecasting. The experimental results show that the proposed method outperforms the state-of-the-art techniques by a considerable margin, both on load profile estimation and peak load forecast problems.

International audience; Finding a set of users that can maximize the spread of information in a social network is an important problem in social media analysis-being a critical part of several realworld applications such as viral marketing, political advertising and epidemiology. Although influence maximization has been studied extensively in the past, the majority of works focus on the algorithmic aspect of the problem, overlooking several practical improvements that can be derived by data-driven observations or the inclusion of machine learning. The main challenges of realistic influence maximization is on the one hand the computational demand of the diffusion models' repetitive simulations, and on the other the accuracy of the estimated influence spread. In this work, we propose CELFIE, an influence maximization method that utilizes learnt influence representations from diffusion cascades to overcome the use of diffusion models. It comprises of two parts. The first is based on INF2VEC, an unsupervised learning model that embeds influence relationships between nodes from a set of diffusion cascades. We create a new version of the model, based on observations from influence analysis on a large scale dataset, to match the scalability needs and the purpose of influence maximization. The second part capitalizes on the learned representations to redefine the traditional live-edge model sampling for the computation of the marginal gain. For evaluation, we apply our method in the Sina Weibo and Microsoft Academic Graph datasets, two large scale networks accompanied by diffusion cascades. We observe that our algorithm outperforms various baseline methods in terms of seed set quality and speed. In addition, the proposed INF2VEC modification for influence maximization provides substantial computational advantages in the price of a minuscule loss in the influence spread.

International audience; Simulated annealing (SA) is a widely used approach to solve global optimization problems in signal processing. The initial non-convex problem is recast as the exploration of a sequence of Boltzmann probability distributions, which are increasingly harder to sample from. They are parametrized by a temperature that is iteratively decreased, following the socalled cooling schedule. Convergence results of SA methods usually require the cooling schedule to be set a priori with slow decay. In this work, we introduce a new SA approach that selects the cooling schedule on the fly. To do so, each Boltzmann distribution is approximated by a proposal density, which is also sequentially adapted. Starting from a variational formulation of the problem of joint temperature and proposal adaptation, we derive an alternating Bregman proximal algorithm to minimize the resulting cost, obtaining the sequence of Boltzmann distributions and proposals. Numerical experiments in an idealized setting illustrate the potential of our method compared with state-of-the-art SA algorithms.

International audience; Electrical motors in industrial and emerging applications such as electrical automotive require high dynamic performance, robustness against parameter variation, and reliability. Recent advances in neural network-based estimators and fault detection techniques rely heavily on accurate sensor information. Due to the extreme operating conditions of electrical motors, there is always a chance of sensor failure which might lead to poor performance in downstream tasks using neural networks. This paper introduces the problem of identifying and recovering sensor faults using generative adversarial networks. We consider sensors monitoring various quantities like currents, voltages, speed, torque, temperature, and vibrations. We introduce fault model for these sensors to simulate training datasets. We use existing GAN based data imputation methods as baseline solutions.

Learning representations of nodes in a low dimensional space is a crucial task with numerous interesting applications in network analysis, including link prediction, node classification, and visualization. Two popular approaches for this problem are matrix factorization and random walk-based models. In this paper, we aim to bring together the best of both worlds, towards learning node representations. In particular, we propose a weighted matrix factorization model that encodes random walk-based information about nodes of the network. The benefit of this novel formulation is that it enables us to utilize kernel functions without realizing the exact proximity matrix so that it enhances the expressiveness of existing matrix decomposition methods with kernels and alleviates their computational complexities. We extend the approach with a multiple kernel learning formulation that provides the flexibility of learning the kernel as the linear combination of a dictionary of kernels in data-driven fashion. We perform an empirical evaluation on real-world networks, showing that the proposed model outperforms baseline node embedding algorithms in downstream machine learning tasks., Comment: This manuscript is an extended version of the previous work entitled "Kernel Node Embeddings" (arXiv:1909.03416), and it has been accepted for publication in IEEE Transactions on Knowledge and Data Engineering

International audience; The stability of neural networks with respect to adversarial perturbations has been extensively studied. One of the main strategies consist of quantifying the Lipschitz regularity of neural networks. In this paper, we introduce a multivariate Lipschitz constant-based stability analysis of fully connected neural networks allowing us to capture the influence of each input or group of inputs on the neural network stability. Our approach relies on a suitable re-normalization of the input space, with the objective to perform a more precise analysis than the one provided by a global Lipschitz constant. We investigate the mathematical properties of the proposed multivariate Lipschitz analysis and show its usefulness in better understanding the sensitivity of the neural network with regard to groups of inputs. We display the results of this analysis by a new representation designed for machine learning practitioners and safety engineers termed as a Lipschitz star. The Lipschitz star is a graphical and practical tool to analyze the sensitivity of a neural network model during its development, with regard to different combinations of inputs. By leveraging this tool, we show that it is possible to build robust-by-design models using spectral normalization techniques for controlling the stability of a neural network, given a safety Lipschitz target. Thanks to our multivariate Lipschitz analysis, we can also measure the efficiency of adversarial training in inference tasks. We perform experiments on various open access tabular datasets, and also on a real Thales Air Mobility industrial application subject to certification requirements.

International audience; Purpose: The objective of our study is to propose fast, cost-effective, convenient, and effective biomarkers using the perfusion parameters from dynamic contrast-enhanced ultrasound (DCE-US) for the evaluation of immune checkpoint inhibitors (ICI) early response. Methods: The retrospective cohort used in this study included 63 patients with metastatic cancer eligible for immunotherapy. DCE-US was performed at baseline, day 8 (D8), and day 21 (D21) after treatment onset. A tumor perfusion curve was modeled on these three dates, and change in the seven perfusion parameters was measured between baseline, D8, and D21. These perfusion parameters were studied to show the impact of their variation on the overall survival (OS). Results: After the removal of missing or suboptimal DCE-US, the Baseline-D8, the Baseline-D21, and the D8-D21 groups included 37, 53, and 33 patients, respectively. A decrease of more than 45% in the area under the perfusion curve (AUC) between baseline and D21 was significantly associated with better OS (p = 0.0114). A decrease of any amount in the AUC between D8 and D21 was also significantly associated with better OS (p = 0.0370). Conclusion: AUC from DCE-US looks to be a promising new biomarker for fast, effective, and convenient immunotherapy response evaluation.

International audience; We demonstrate the interest of multiphoton microscopy (MPM) for imaging bacteria without any labelling process. Six families of bacteria are tested: Escherichia coli, Staphylococcus epidermidis, Proteus vulgaris, Pseudomonas fluorescens, Bacillus subtilis and Clostridium perfringens. For each of these bacteria, the image of a cell is recorded through a multiphoton microscope thus revealing the 3D shape of these bacteria. For the first time, the images of such bacteria are recorded without any labelling solution. A protocol of controlling the image produced is led thanks to a standard staining protocol with carboxy fluorescein diacetate (CFDA) for E. Coli and Staphylococcus epidermidis. Similar object shapes with or without labelling are produced, thus validating the label-free images generated by MPM. Then, the two-photon excitation spectra are measured for each of these bacteriaand the emission spectra delivered by E. Coli and Bacillus subtilis are shown. The origin of the two-photon fluorescence (TPF) emission of the bacteria thanks to the nonlinear imaging solution is discussed regarding to the TPF excitation and emission spectra of metabolic indicators.

MotivationThe cellular system of a living organism is composed of interacting bio-molecules that control cellular processes at multiple levels. Their correspondences are represented by tightly regulated molecular networks. The increase of omics technologies has favored the generation of large-scale disparate data and the consequent demand for simultaneously using molecular and functional interaction networks: gene co-expression, protein–protein interaction (PPI), genetic interaction and metabolic networks. They are rich sources of information at different molecular levels, and their effective integration is essential to understand cell functioning and their building blocks (proteins). Therefore, it is necessary to obtain informative representations of proteins and their proximity, that are not fully captured by features extracted directly from a single informational level. We propose BraneMF, a novel random walk-based matrix factorization method for learning node representation in a multilayer network, with application to omics data integration.ResultsWe test BraneMF with PPI networks of Saccharomyces cerevisiae, a well-studied yeast model organism. We demonstrate the applicability of the learned features for essential multi-omics inference tasks: clustering, function and PPI prediction. We compare it to the state-of-the-art integration methods for multilayer networks. BraneMF outperforms baseline methods by achieving high prediction scores for a variety of downstream tasks. The robustness of results is assessed by an extensive parameter sensitivity analysis.Availability and implementationBraneMF’s code is freely available at: https://github.com/Surabhivj/BraneMF, along with datasets, embeddings and result files.Supplementary informationSupplementary data are available at Bioinformatics online.

Bayesian neural networks (BNNs) have received an increased interest in the last years. In BNNs, a complete posterior distribution of the unknown weight and bias parameters of the network is produced during the training stage. This probabilistic estimation offers several advantages with respect to point-wise estimates, in particular, the ability to provide uncertainty quantification when predicting new data. This feature inherent to the Bayesian paradigm, is useful in countless machine learning applications. It is particularly appealing in areas where decision-making has a crucial impact, such as medical healthcare or autonomous driving. The main challenge of BNNs is the computational cost of the training procedure since Bayesian techniques often face a severe curse of dimensionality. Adaptive importance sampling (AIS) is one of the most prominent Monte Carlo methodologies benefiting from sounded convergence guarantees and ease for adaptation. This work aims to show that AIS constitutes a successful approach for designing BNNs. More precisely, we propose a novel algorithm PMCnet that includes an efficient adaptation mechanism, exploiting geometric information on the complex (often multimodal) posterior distribution. Numerical results illustrate the excellent performance and the improved exploration capabilities of the proposed method for both shallow and deep neural networks.

The joint problem of reconstruction/feature extraction is a challenging task in image processing. It consists in performing, in a joint manner, the restoration of an image and the extraction of its features. In this work, we firstly propose a novel nonsmooth and nonconvex variational formulation of the problem. For this purpose, we introduce a versatile generalised Gaussian prior whose parameters, including its exponents, are space-variant. Secondly, we design an alternating proximal-based optimisation algorithm that efficiently exploits the structure of the proposed nonconvex objective function. We also analyze the convergence of this algorithm. As shown in numerical experiments conducted on joint segmentation/deblurring tasks, the proposed method provides high-quality results.

We study the variational inference problem of minimizing a regularized Rényi divergence over an exponential family, and propose a relaxed moment-matching algorithm, which includes a proximal-like step. Using the information-geometric link between Bregman divergences and the Kullback-Leibler divergence, this algorithm is shown to be equivalent to a Bregman proximal gradient algorithm. This novel perspective allows us to exploit the geometry of our approximate model while using stochastic black-box updates. We use this point of view to prove strong convergence guarantees including monotonic decrease of the objective, convergence to a stationary point or to the minimizer, and geometric convergence rates. These new theoretical insights lead to a versatile, robust, and competitive method, as illustrated by numerical experiments.

International audience; In Drug-Drug-Interaction (DDI), the task is to predict the (adverse) effect of administering two drugs simultaneously. Currently, the techniques proposed in this direction are generally based on either shallow learning paradigms like Random Decision Forest (RDF), Logistic Regression (LR), Support Vector Machines (SVM), etc., or deep Convolutional Neural Networks (CNNs). However, specific works combine traditional machine learning (ML) algorithms such as RDF, LR, SVM, and deep learning paradigms such as CNNs in a piecemeal fashion which might not be optimal. Hence, the present work proposes a framework that presents a joint end-to-end solution. We propose a Siamese-like architecture with two processing channels' networks based on deep convolutional transform learning. Common fused representations as well as channel-wise representations are learnt, in addition with the transform across them. The final representation is passed to a decision forest to give final predictions. The proposed method is thus a supervised end-to-end multi-channel fusion framework that (i) learns unique and interpretable filters in contrast with CNNs, and (ii) jointly learns and optimizes decision forest in contrast with state-of-the-art piecemeal approach. We apply this technique to identify DDIs among 1059 drugs from the DrugBank database showing superiority of our method compared to the state-of-the-art(s).

Graph Neural Networks achieve state-of-the-art performance on a plethora of graph classification tasks, especially due to pooling operators, which aggregate learned node embeddings hierarchically into a final graph representation. However, they are not only questioned by recent work showing on par performance with random pooling, but also ignore completely higher-order connectivity patterns. To tackle this issue, we propose HoscPool, a clustering-based graph pooling operator that captures higher-order information hierarchically, leading to richer graph representations. In fact, we learn a probabilistic cluster assignment matrix end-to-end by minimising relaxed formulations of motif spectral clustering in our objective function, and we then extend it to a pooling operator. We evaluate HoscPool on graph classification tasks and its clustering component on graphs with ground-truth community structure, achieving best performance. Lastly, we provide a deep empirical analysis of pooling operators' inner functioning., Comment: CIKM 2022

The core of many approaches for the resolution of variational inverse problems arising in signal and image processing consists of promoting the sought solution to have a sparse representation in a well-suited space. A crucial task in this context is the choice of a good sparsity prior that can ensure a good trade-off between the quality of the solution and the resulting computational cost. The recently introduced Convex-Non-Convex (CNC) strategy appears as a great compromise, as it combines the high qualitative performance of non-convex sparsity-promoting functions with the convenience of dealing with convex optimization problems. This work proposes a new variational formulation to implement CNC approach in the context of image denoising. By suitably exploiting duality properties, our formulation allows to encompass sophisticated directional total variation (DTV) priors. We additionally propose an efficient optimisation strategy for the resulting convex minimisation problem. We illustrate on numerical examples the good performance of the resulting CNC-DTV method, when compared to the standard convex total variation denoiser., Accepted for EUSIPCO 2022 - 30th European Signal Processing Conference, Aug 2022, Belgrade, Serbia

International audience; Electrical motors are widely used in industrial and emerging applications such as electrical automotive. Industrial 4.0 has led to the usage of neural networks for electrical motor tasks like fault detection, monitoring, and control of electrical motors. The growing increase of neural networks in safety-critical systems requires an in-depth analysis of their robustness and stability. This paper studies the robustness of neural networks used in time-series tasks like system modeling, signal denoising, speed-torque estimation, temperature estimation, and fault detection. The dataset collected for these problems has all types of noise from the operating environment, sensors, and the system itself. This affects the performance of different network architectures during training and inference. We train and analyze under perturbations several different architectures that range from simple linear, convolutional and sequential networks to complex networks like 1D ResNet and Transformers.

Gliomas are among the most common types of central nervous system (CNS) tumors. A prompt diagnosis of the glioma subtype is crucial to estimate the prognosis and personalize the treatment strategy. The objective of this study was to develop a radiomics pipeline based on the clinical Magnetic Resonance Imaging (MRI) scans to noninvasively predict the glioma subtype, as defined based on the tumor grade, isocitrate dehydrogenase (IDH) mutation status, and 1p/19q codeletion status. A total of 212 patients from the public retrospective The Cancer Genome Atlas Low Grade Glioma (TCGA-LGG) and The Cancer Genome Atlas Glioblastoma Multiforme (TCGA-GBM) datasets were used for the experiments and analyses. Different settings in the radiomics pipeline were investigated to improve the classification, including the Z-score normalization, the feature extraction strategy, the image filter applied to the MRI images, the introduction of clinical information, ComBat harmonization, the classifier chain strategy, etc. Based on numerous experiments, we finally reached an optimal pipeline for classifying the glioma tumors. We then tested this final radiomics pipeline on the hold-out test data with 51 randomly sampled random seeds for reliable and robust conclusions. The results showed that, after tuning the radiomics pipeline, the mean AUC improved from 0.8935 (±0.0351) to 0.9319 (±0.0386), from 0.8676 (±0.0421) to 0.9283 (±0.0333), and from 0.6473 (±0.1074) to 0.8196 (±0.0702) in the test data for predicting the tumor grade, IDH mutation, and 1p/19q codeletion status, respectively. The mean accuracy for predicting the five glioma subtypes also improved from 0.5772 (±0.0816) to 0.6716 (±0.0655). Finally, we analyzed the characteristics of the radiomic features that best distinguished the glioma grade, the IDH mutation, and the 1p/19q codeletion status, respectively. Apart from the promising prediction of the glioma subtype, this study also provides a better understanding of the radiomics model development and interpretability. The results in this paper are replicable with our python codes publicly available in github.

This paper addresses the problem of image reconstruction for region-of-interest (ROI) computed tomography (CT). While model-based iterative methods can be used for such a problem, their practicability is often limited due to tedious parameterization and slow convergence. In addition, inadequate solutions can be obtained when the retained priors do not perfectly fit the solution space. Deep learning methods offer an alternative approach that is fast, leverages information from large data sets, and thus can reach high reconstruction quality. However, these methods usually rely on black boxes not accounting for the physics of the imaging system, and their lack of interpretability is often deplored. At the crossroads of both methods, unfolded deep learning techniques have been recently proposed. They incorporate the physics of the model and iterative optimization algorithms into a neural network design, leading to superior performance in various applications. This paper introduces a novel, unfolded deep learning approach called U-RDBFB designed for ROI CT reconstruction from limited data. Few-view truncated data are efficiently handled thanks to a robust non-convex data fidelity function combined with sparsityinducing regularization functions. Iterations of a block dual forward-backward (DBFB) algorithm, embedded in an iterative reweighted scheme, are then unrolled over a neural network architecture, allowing the learning of various parameters in a supervised manner. Our experiments show an improvement over various state-of-the-art methods, including model-based iterative schemes, deep learning architectures, and deep unfolding methods.