arXiv subjects

Pascal Frossard

Publications and source records attributed to Pascal Frossard.

At least 145 records · Page 8Linked to original sources

arXiv2019

Forward-Backward Splitting for Optimal Transport based Problems

Optimal transport aims to estimate a transportation plan that minimizes a displacement cost. This is realized by optimizing the scalar product between the sought plan and the given cost, over the space of doubly stochastic matrices. When the entropy regularization is added to the problem, the transportation plan can be efficiently computed with the Sinkhorn algorithm. Thanks to this breakthrough, optimal transport has been progressively extended to machine learning and statistical inference by introducing additional application-specific terms in the problem formulation. It is however challenging to design efficient optimization algorithms for optimal transport based extensions. To overcome this limitation, we devise a general forward-backward splitting algorithm based on Bregman distances for solving a wide range of optimization problems involving a differentiable function with Lipschitz-continuous gradient and a doubly stochastic constraint. We illustrate the efficiency of our approach in the context of continuous domain adaptation. Experiments show that the proposed method leads to a significant improvement in terms of speed and performance with respect to the state of the art for domain adaptation on a continually rotating distribution coming from the standard two moon dataset.

stat.ML↗

arXiv2019

Optimized Quantization in Distributed Graph Signal Filtering

Distributed graph signal processing algorithms require the network nodes to communicate by exchanging messages in order to achieve a common objective. These messages have a finite precision in realistic networks, which may necessitate to implement message quantization. Quantization, in turn, may generate distortion and performance penalty in the distributed processing tasks. This paper proposes a novel method for distributed graph filtering that minimizes the error due to message quantization without compromising the communication costs. It first bounds the exchanged messages and then allocates a limited bit budget in an optimized way to the different messages and network nodes. In particular, our novel quantization algorithm adapts to both the network topology and the message importance in a distributed processing task. Our results show that the proposed method is effective in minimizing the error due to quantization and that it permits to outperform baseline distributed algorithms when the bit budget is limited. They further show that errors produced in nodes with high eccentricity or in the first steps of the distributed algorithm contribute more to the global error. Also, sparse and irregular graphs require more irregular bit distribution. Our method provides one of the first quantization solutions for distributed graph processing, which is able to adapt to the target task, the graph properties and the communication constraints.

eess.SP↗

arXiv2019

Graph Transform Optimization with Application to Image Compression

In this paper, we propose a new graph-based transform and illustrate its potential application to signal compression. Our approach relies on the careful design of a graph that optimizes the overall rate-distortion performance through an effective graph-based transform. We introduce a novel graph estimation algorithm, which uncovers the connectivities between the graph signal values by taking into consideration the coding of both the signal and the graph topology in rate-distortion terms. In particular, we introduce a novel coding solution for the graph by treating the edge weights as another graph signal that lies on the dual graph. Then, the cost of the graph description is introduced in the optimization problem by minimizing the sparsity of the coefficients of its graph Fourier transform (GFT) on the dual graph. In this way, we obtain a convex optimization problem whose solution defines an efficient transform coding strategy. The proposed technique is a general framework that can be applied to different types of signals, and we show two possible application fields, namely natural image coding and piecewise smooth image coding. The experimental results show that the proposed graph-based transform outperforms classical fixed transforms such as DCT for both natural and piecewise smooth images. In the case of depth map coding, the obtained results are even comparable to the state-of-the-art graph-based coding method, that are specifically designed for depth map images.

cs.IT↗

arXiv2019

SparseFool: a few pixels make a big difference

Deep Neural Networks have achieved extraordinary results on image classification tasks, but have been shown to be vulnerable to attacks with carefully crafted perturbations of the input data. Although most attacks usually change values of many image's pixels, it has been shown that deep networks are also vulnerable to sparse alterations of the input. However, no computationally efficient method has been proposed to compute sparse perturbations. In this paper, we exploit the low mean curvature of the decision boundary, and propose SparseFool, a geometry inspired sparse attack that controls the sparsity of the perturbations. Extensive evaluations show that our approach computes sparse perturbations very fast, and scales efficiently to high dimensional data. We further analyze the transferability and the visual effects of the perturbations, and show the existence of shared semantic information across the images and the networks. Finally, we show that adversarial training can only slightly improve the robustness against sparse additive perturbations computed with SparseFool.

cs.CV↗

arXiv2019

Learning graphs from data: A signal representation perspective

The construction of a meaningful graph topology plays a crucial role in the effective representation, processing, analysis and visualization of structured data. When a natural choice of the graph is not readily available from the data sets, it is thus desirable to infer or learn a graph topology from the data. In this tutorial overview, we survey solutions to the problem of graph learning, including classical viewpoints from statistics and physics, and more recent approaches that adopt a graph signal processing (GSP) perspective. We further emphasize the conceptual similarities and differences between classical and GSP-based graph inference methods, and highlight the potential advantage of the latter in a number of theoretical and practical scenarios. We conclude with several open issues and challenges that are keys to the design of future signal processing and machine learning algorithms for learning graphs from data.

cs.LG↗

arXiv2019

Stochastic Gradient Descent for Spectral Embedding with Implicit Orthogonality Constraint

In this paper, we propose a scalable algorithm for spectral embedding. The latter is a standard tool for graph clustering. However, its computational bottleneck is the eigendecomposition of the graph Laplacian matrix, which prevents its application to large-scale graphs. Our contribution consists of reformulating spectral embedding so that it can be solved via stochastic optimization. The idea is to replace the orthogonality constraint with an orthogonalization matrix injected directly into the criterion. As the gradient can be computed through a Cholesky factorization, our reformulation allows us to develop an efficient algorithm based on mini-batch gradient descent. Experimental results, both on synthetic and real data, confirm the efficiency of the proposed method in term of execution speed with respect to similar existing techniques.

cs.LG↗

arXiv2019

A geometry-inspired decision-based attack

Deep neural networks have recently achieved tremendous success in image classification. Recent studies have however shown that they are easily misled into incorrect classification decisions by adversarial examples. Adversaries can even craft attacks by querying the model in black-box settings, where no information about the model is released except its final decision. Such decision-based attacks usually require lots of queries, while real-world image recognition systems might actually restrict the number of queries. In this paper, we propose qFool, a novel decision-based attack algorithm that can generate adversarial examples using a small number of queries. The qFool method can drastically reduce the number of queries compared to previous decision-based attacks while reaching the same quality of adversarial examples. We also enhance our method by constraining adversarial perturbations in low-frequency subspace, which can make qFool even more computationally efficient. Altogether, we manage to fool commercial image recognition systems with a small number of queries, which demonstrates the actual effectiveness of our new algorithm in practice.

cs.LG↗

arXiv2019

Graph heat mixture model learning

Graph inference methods have recently attracted a great interest from the scientific community, due to the large value they bring in data interpretation and analysis. However, most of the available state-of-the-art methods focus on scenarios where all available data can be explained through the same graph, or groups corresponding to each graph are known a priori. In this paper, we argue that this is not always realistic and we introduce a generative model for mixed signals following a heat diffusion process on multiple graphs. We propose an expectation-maximisation algorithm that can successfully separate signals into corresponding groups, and infer multiple graphs that govern their behaviour. We demonstrate the benefits of our method on both synthetic and real data.

cs.LG↗

arXiv2019

Visual Distortions in 360-degree Videos

Omnidirectional (or 360-degree) images and videos are emergent signals in many areas such as robotics and virtual/augmented reality. In particular, for virtual reality, they allow an immersive experience in which the user is provided with a 360-degree field of view and can navigate throughout a scene, e.g., through the use of Head Mounted Displays. Since it represents the full 360-degree field of view from one point of the scene, omnidirectional content is naturally represented as spherical visual signals. Current approaches for capturing, processing, delivering, and displaying 360-degree content, however, present many open technical challenges and introduce several types of distortions in these visual signals. Some of the distortions are specific to the nature of 360-degree images, and often different from those encountered in the classical image communication framework. This paper provides a first comprehensive review of the most common visual distortions that alter 360-degree signals undergoing state of the art processing in common applications. While their impact on viewers' visual perception and on the immersive experience at large is still unknown ---thus, it stays an open research topic--- this review serves the purpose of identifying the main causes of visual distortions in the end-to-end 360-degree content distribution pipeline. It is essential as a basis for benchmarking different processing techniques, allowing the effective design of new algorithms and applications. It is also necessary to the deployment of proper psychovisual studies to characterise the human perception of these new images in interactive and immersive applications.

cs.MM↗

arXiv2018

Graph Signal Representation with Wasserstein Barycenters

In many applications signals reside on the vertices of weighted graphs. Thus, there is the need to learn low dimensional representations for graph signals that will allow for data analysis and interpretation. Existing unsupervised dimensionality reduction methods for graph signals have focused on dictionary learning. In these works the graph is taken into consideration by imposing a structure or a parametrization on the dictionary and the signals are represented as linear combinations of the atoms in the dictionary. However, the assumption that graph signals can be represented using linear combinations of atoms is not always appropriate. In this paper we propose a novel representation framework based on non-linear and geometry-aware combinations of graph signals by leveraging the mathematical theory of Optimal Transport. We represent graph signals as Wasserstein barycenters and demonstrate through our experiments the potential of our proposed framework for low-dimensional graph signal representation.

eess.SP↗

arXiv2018

Robustness via curvature regularization, and vice versa

State-of-the-art classifiers have been shown to be largely vulnerable to adversarial perturbations. One of the most effective strategies to improve robustness is adversarial training. In this paper, we investigate the effect of adversarial training on the geometry of the classification landscape and decision boundaries. We show in particular that adversarial training leads to a significant decrease in the curvature of the loss surface with respect to inputs, leading to a drastically more "linear" behaviour of the network. Using a locally quadratic approximation, we provide theoretical evidence on the existence of a strong relation between large robustness and small curvature. To further show the importance of reduced curvature for improving the robustness, we propose a new regularizer that directly minimizes curvature of the loss surface, and leads to adversarial robustness that is on par with adversarial training. Besides being a more efficient and principled alternative to adversarial training, the proposed regularizer confirms our claims on the importance of exhibiting quasi-linear behavior in the vicinity of data points in order to achieve robustness.

cs.LG↗

arXiv2018

Kernel Regression for Graph Signal Prediction in Presence of Sparse Noise

In presence of sparse noise we propose kernel regression for predicting output vectors which are smooth over a given graph. Sparse noise models the training outputs being corrupted either with missing samples or large perturbations. The presence of sparse noise is handled using appropriate use of $\ell_1$-norm along-with use of $\ell_2$-norm in a convex cost function. For optimization of the cost function, we propose an iteratively reweighted least-squares (IRLS) approach that is suitable for kernel substitution or kernel trick due to availability of a closed form solution. Simulations using real-world temperature data show efficacy of our proposed method, mainly for limited-size training datasets.

stat.ML↗

arXiv2018

Convolutional neural networks on irregular domains based on approximate vertex-domain translations

We propose a generalization of convolutional neural networks (CNNs) to irregular domains, through the use of a translation operator on a graph structure. In regular settings such as images, convolutional layers are designed by translating a convolutional kernel over all pixels, thus enforcing translation equivariance. In the case of general graphs however, translation is not a well-defined operation, which makes shifting a convolutional kernel not straightforward. In this article, we introduce a methodology to allow the design of convolutional layers that are adapted to signals evolving on irregular topologies, even in the absence of a natural translation. Using the designed layers, we build a CNN that we train using the initial set of signals. Contrary to other approaches that aim at extending CNNs to irregular domains, we incorporate the classical settings of CNNs for 2D signals as a particular case of our approach. Designing convolutional layers in the vertex domain directly implies weight sharing, which in other approaches is generally estimated a posteriori using heuristics.

cs.DM↗

arXiv2018

Supervised Linear Regression for Graph Learning from Graph Signals

We propose a supervised learning approach for predicting an underlying graph from a set of graph signals. Our approach is based on linear regression. In the linear regression model, we predict edge-weights of a graph as the output, given a set of signal values on nodes of the graph as the input. We solve for the optimal regression coefficients using a relevant optimization problem that is convex and uses a graph-Laplacian based regularization. The regularization helps to promote a specific graph spectral profile of the graph signals. Simulation experiments demonstrate that our approach predicts well even in presence of outliers in input data.

cs.IT↗

arXiv2018

Isometric Transformation Invariant Graph-based Deep Neural Network

Learning transformation invariant representations of visual data is an important problem in computer vision. Deep convolutional networks have demonstrated remarkable results for image and video classification tasks. However, they have achieved only limited success in the classification of images that undergo geometric transformations. In this work we present a novel Transformation Invariant Graph-based Network (TIGraNet), which learns graph-based features that are inherently invariant to isometric transformations such as rotation and translation of input images. In particular, images are represented as signals on graphs, which permits to replace classical convolution and pooling layers in deep networks with graph spectral convolution and dynamic graph pooling layers that together contribute to invariance to isometric transformation. Our experiments show high performance on rotated and translated images from the test set compared to classical architectures that are very sensitive to transformations in the data. The inherent invariance properties of our framework provide key advantages, such as increased resiliency to data variability and sustained performance with limited training sets. Our code is available online.

cs.CV↗

arXiv2018

Online Resource Inference in Network Utility Maximization Problems

The amount of transmitted data in computer networks is expected to grow considerably in the future, putting more and more pressure on the network infrastructures. In order to guarantee a good service, it then becomes fundamental to use the network resources efficiently. Network Utility Maximization (NUM) provides a framework to optimize the rate allocation when network resources are limited. Unfortunately, in the scenario where the amount of available resources is not known a priori, classical NUM solving methods do not offer a viable solution. To overcome this limitation we design an overlay rate allocation scheme that attempts to infer the actual amount of available network resources while coordinating the users rate allocation. Due to the general and complex model assumed for the congestion measurements, a passive learning of the available resources would not lead to satisfying performance. The coordination scheme must then perform active learning in order to speed up the resources estimation and quickly increase the system performance. By adopting an optimal learning formulation we are able to balance the tradeoff between an accurate estimation, and an effective resources exploitation in order to maximize the long term quality of the service delivered to the users.

cs.NI↗

arXiv2018

Graph Signal Processing: Overview, Challenges and Applications

Research in Graph Signal Processing (GSP) aims to develop tools for processing data defined on irregular graph domains. In this paper we first provide an overview of core ideas in GSP and their connection to conventional digital signal processing. We then summarize recent developments in developing basic GSP tools, including methods for sampling, filtering or graph learning. Next, we review progress in several application areas using GSP, including processing and analysis of sensor network data, biological data, and applications to image processing and machine learning. We finish by providing a brief historical perspective to highlight how concepts recently developed in GSP build on top of prior research in other areas.

eess.SP↗

arXiv2018

Adaptive Streaming in Interactive Multiview Video Systems

Multiview applications endow final users with the possibility to freely navigate within 3D scenes with minimum-delay. A real feeling of scene navigation is enabled by transmitting multiple high-quality camera views, which can be used to synthesize additional virtual views to offer a smooth navigation. However, when network resources are limited, not all camera views can be sent at high quality. It is therefore important, yet challenging, to find the right tradeoff between coding artifacts (reducing the quality of camera views) and virtual synthesis artifacts (reducing the number of camera views sent to users). To this aim, we propose an optimal transmission strategy for interactive multiview HTTP adaptive streaming (HAS). We propose a problem formulation to select the optimal set of camera views that the client requests for downloading, such that the navigation quality experienced by the user is optimized while the bandwidth constraints are satisfied. We show that our optimization problem is NP-hard, and we therefore develop an optimal solution based on the dynamic programming algorithm with polynomial time complexity. To further simplify the deployment, we present a suboptimal greedy algorithm with effective performance and lower complexity. The proposed controller is evaluated in theoretical and realistic settings characterized by realistic network statistics estimation, buffer management and server-side representation optimization. Simulation results show significant improvement in terms of navigation quality compared with alternative baseline multiview adaptation logic solutions.

cs.MM↗