arXiv subjects

Guillaume Bouchard

Publications and source records attributed to Guillaume Bouchard.

18 recordsLinked to original sources

arXiv2024

Perfectly Matched Layer implementation for E-H fields and Complex Wave Envelope propagation in the Smilei PIC code

The design of absorbing boundary conditions (ABC) in a numerical simulation is a challenging task. In the best cases, spurious reflections remain for some angles of incidence or at certain wave lengths. In the worst, ABC are not even possible for the set of equations and/or numerical schemes used in the simulation and reflections can not be avoided at all. Perflectly Matched Layer (PML) are layers of absorbing medium which can be added at the simulation edges in order to significantly damp both outgoing and reflected waves, thus effectively playing the role of an ABC. They are able to absorb waves and prevent reflections for all angles and frequencies at a modest computational cost. It increases the simulation accuracy and negates the need of oversizing the simulation usually imposed by ABC and leading to a waste of computational resources and power. PML for finite-difference time-domain (FDTD) schemes in Particle-In-cell (PIC) codes are presented for both Maxwell's equations and, for the first time, the envelope wave equation. Being of the second order, the latter requires significant evolutions with respect to the former. It applies in particular to simulations of lasers propagating in plasmas using the reduced Complex Envelope model. The implementation is done in the open source code Smilei for both Cartesian and azimuthal modes (AM) decomposition geometries.

physics.comp-ph↗

arXiv2023

Detecting Harmful Content On Online Platforms: What Platforms Need Vs. Where Research Efforts Go

The proliferation of harmful content on online platforms is a major societal problem, which comes in many different forms including hate speech, offensive language, bullying and harassment, misinformation, spam, violence, graphic content, sexual abuse, self harm, and many other. Online platforms seek to moderate such content to limit societal harm, to comply with legislation, and to create a more inclusive environment for their users. Researchers have developed different methods for automatically detecting harmful content, often focusing on specific sub-problems or on narrow communities, as what is considered harmful often depends on the platform and on the context. We argue that there is currently a dichotomy between what types of harmful content online platforms seek to curb, and what research efforts there are to automatically detect such content. We thus survey existing methods as well as content moderation policies by online platforms in this light and we suggest directions for future work.

cs.CL↗

arXiv2022

A Task Programming Implementation for the Particle in Cell Code Smilei

An implementation of the electromagnetic Particle in Cell loop in the code Smilei using task programming is presented. Through OpenMP, the macro-particles operations are formulated in terms of tasks. This formulation allows asynchronous execution respecting the data dependencies of the macro-particle operations, the most time-consuming part of the code in simulations of interest for plasma physics. Through some benchmarks it is shown that this formulation can help mitigating the load imbalance of these operations at the OpenMP thread level. The improvements in strong scaling for load-imbalanced physical cases are discussed.

cs.DC↗

arXiv2018

Interpretation of Natural Language Rules in Conversational Machine Reading

Most work in machine reading focuses on question answering problems where the answer is directly expressed in the text to read. However, many real-world question answering problems require the reading of text not because it contains the literal answer, but because it contains a recipe to derive an answer together with the reader's background knowledge. One example is the task of interpreting regulations to answer "Can I...?" or "Do I have to...?" questions such as "I am working in Canada. Do I have to carry on paying UK National Insurance?" after reading a UK government website about this topic. This task requires both the interpretation of rules and the application of background knowledge. It is further complicated due to the fact that, in practice, most questions are underspecified, and a human assistant will regularly have to ask clarification questions such as "How long have you been working abroad?" when the answer cannot be directly derived from the question and text. In this paper, we formalise this task and develop a crowd-sourcing strategy to collect 32k task instances based on real-world rules and crowd-generated questions and scenarios. We analyse the challenges of this task and assess its difficulty by evaluating the performance of rule-based and machine-learning baselines. We observe promising results when no background knowledge is necessary, and substantial room for improvement whenever background knowledge is needed.

cs.CL↗

arXiv2017

Knowledge Graph Completion via Complex Tensor Factorization

In statistical relational learning, knowledge graph completion deals with automatically understanding the structure of large knowledge graphs---labeled directed graphs---and predicting missing relationships---labeled edges. State-of-the-art embedding models propose different trade-offs between modeling expressiveness, and time and space complexity. We reconcile both expressiveness and complexity through the use of complex-valued embeddings and explore the link between such complex-valued embeddings and unitary diagonalization. We corroborate our approach theoretically and show that all real square matrices---thus all possible relation/adjacency matrices---are the real part of some unitarily diagonalizable matrix. This results opens the door to a lot of other applications of square matrices factorization. Our approach based on complex embeddings is arguably simple, as it only involves a Hermitian dot product, the complex counterpart of the standard dot product between real vectors, whereas other methods resort to more and more complicated composition functions to increase their expressiveness. The proposed complex embeddings are scalable to large data sets as it remains linear in both space and time, while consistently outperforming alternative approaches on standard link prediction benchmarks.

cs.AI↗

arXiv2017

On Inductive Abilities of Latent Factor Models for Relational Learning

Latent factor models are increasingly popular for modeling multi-relational knowledge graphs. By their vectorial nature, it is not only hard to interpret why this class of models works so well, but also to understand where they fail and how they might be improved. We conduct an experimental survey of state-of-the-art models, not towards a purely comparative end, but as a means to get insight about their inductive abilities. To assess the strengths and weaknesses of each model, we create simple tasks that exhibit first, atomic properties of binary relations, and then, common inter-relational inference through synthetic genealogies. Based on these experimental results, we propose new research directions to improve on existing models.

cs.LG↗

arXiv2016

SentiHood: Targeted Aspect Based Sentiment Analysis Dataset for Urban Neighbourhoods

In this paper, we introduce the task of targeted aspect-based sentiment analysis. The goal is to extract fine-grained information with respect to entities mentioned in user comments. This work extends both aspect-based sentiment analysis that assumes a single entity per document and targeted sentiment analysis that assumes a single sentiment towards a target entity. In particular, we identify the sentiment towards each aspect of one or more entities. As a testbed for this task, we introduce the SentiHood dataset, extracted from a question answering (QA) platform where urban neighbourhoods are discussed by users. In this context units of text often mention several aspects of one or more neighbourhoods. This is the first time that a generic social media platform in this case a QA platform, is used for fine-grained opinion mining. Text coming from QA platforms is far less constrained compared to text from review specific platforms which current datasets are based on. We develop several strong baselines, relying on logistic regression and state-of-the-art recurrent neural networks.

cs.CL↗

arXiv2016

Complex Embeddings for Simple Link Prediction

In statistical relational learning, the link prediction problem is key to automatically understand the structure of large knowledge bases. As in previous studies, we propose to solve this problem through latent factorization. However, here we make use of complex valued embeddings. The composition of complex embeddings can handle a large variety of binary relations, among them symmetric and antisymmetric relations. Compared to state-of-the-art models such as Neural Tensor Network and Holographic Embeddings, our approach based on complex embeddings is arguably simpler, as it only uses the Hermitian dot product, the complex counterpart of the standard dot product between real vectors. Our approach is scalable to large datasets as it remains linear in both space and time, while consistently outperforming alternative approaches on standard link prediction benchmarks.

cs.AI↗

arXiv2016

Decomposing Real Square Matrices via Unitary Diagonalization

Diagonalization, or eigenvalue decomposition, is very useful in many areas of applied mathematics, including signal processing and quantum physics. Matrix decomposition is also a useful tool for approximating matrices as the product of a matrix and its transpose, which relates to unitary diagonalization. As stated by the spectral theorem, only normal matrices are unitarily diagonalizable. However we show that all real square matrices are the real part of some unitarily diagonalizable matrix.

math.SP↗

arXiv2016

A Factorization Machine Framework for Testing Bigram Embeddings in Knowledgebase Completion

Embedding-based Knowledge Base Completion models have so far mostly combined distributed representations of individual entities or relations to compute truth scores of missing links. Facts can however also be represented using pairwise embeddings, i.e. embeddings for pairs of entities and relations. In this paper we explore such bigram embeddings with a flexible Factorization Machine model and several ablations from it. We investigate the relevance of various bigram types on the fb15k237 dataset and find relative improvements compared to a compositional model.

cs.CL↗

arXiv2015

Online Learning to Sample

Stochastic Gradient Descent (SGD) is one of the most widely used techniques for online optimization in machine learning. In this work, we accelerate SGD by adaptively learning how to sample the most useful training examples at each time step. First, we show that SGD can be used to learn the best possible sampling distribution of an importance sampling estimator. Second, we show that the sampling distribution of an SGD algorithm can be estimated online by incrementally minimizing the variance of the gradient. The resulting algorithm - called Adaptive Weighted SGD (AW-SGD) - maintains a set of parameters to optimize, as well as a set of parameters to sample learning examples. We show that AWSGD yields faster convergence in three different applications: (i) image classification with deep features, where the sampling of images depends on their labels, (ii) matrix factorization, where rows and columns are not sampled uniformly, and (iii) reinforcement learning, where the optimized and exploration policies are estimated at the same time, where our approach corresponds to an off-policy gradient algorithm.

cs.LG↗

arXiv2015

Approximate Inference with the Variational Holder Bound

We introduce the Variational Holder (VH) bound as an alternative to Variational Bayes (VB) for approximate Bayesian inference. Unlike VB which typically involves maximization of a non-convex lower bound with respect to the variational parameters, the VH bound involves minimization of a convex upper bound to the intractable integral with respect to the variational parameters. Minimization of the VH bound is a convex optimization problem; hence the VH method can be applied using off-the-shelf convex optimization algorithms and the approximation error of the VH bound can also be analyzed using tools from convex optimization literature. We present experiments on the task of integrating a truncated multivariate Gaussian distribution and compare our method to VB, EP and a state-of-the-art numerical integration method for this problem.

stat.ML↗

arXiv2014

Overlapping Trace Norms in Multi-View Learning

Multi-view learning leverages correlations between different sources of data to make predictions in one view based on observations in another view. A popular approach is to assume that, both, the correlations between the views and the view-specific covariances have a low-rank structure, leading to inter-battery factor analysis, a model closely related to canonical correlation analysis. We propose a convex relaxation of this model using structured norm regularization. Further, we extend the convex formulation to a robust version by adding an l1-penalized matrix to our estimator, similarly to convex robust PCA. We develop and compare scalable algorithms for several convex multi-view models. We show experimentally that the view-specific correlations are improving data imputation performances, as well as labeling accuracy in real-world multi-label prediction tasks.

cs.LG↗

arXiv2014

Group-sparse Embeddings in Collective Matrix Factorization

CMF is a technique for simultaneously learning low-rank representations based on a collection of matrices with shared entities. A typical example is the joint modeling of user-item, item-property, and user-feature matrices in a recommender system. The key idea in CMF is that the embeddings are shared across the matrices, which enables transferring information between them. The existing solutions, however, break down when the individual matrices have low-rank structure not shared with others. In this work we present a novel CMF solution that allows each of the matrices to have a separate low-rank structure that is independent of the other matrices, as well as structures that are shared only by a subset of them. We compare MAP and variational Bayesian solutions based on alternating optimization algorithms and show that the model automatically infers the nature of each factor using group-wise sparsity. Our approach supports in a principled way continuous, binary and count observations and is efficient for sparse matrices involving missing data. We illustrate the solution on a number of examples, focusing in particular on an interesting use-case of augmented multi-view learning.

stat.ML↗

arXiv2013

Testing for Associations between Loci and Environmental Gradients Using Latent Factor Mixed Models

Adaptation to local environments often occurs through natural selection acting on a large number of loci, each having a weak phenotypic effect. One way to detect these loci is to identify genetic polymorphisms that exhibit high correlation with environmental variables used as proxies for ecological pressures. Here, we propose new algorithms based on population genetics, ecological modeling, and statistical learning techniques to screen genomes for signatures of local adaptation. Implemented in the computer program "latent factor mixed model" (LFMM), these algorithms employ an approach in which population structure is introduced using unobserved variables. These fast and computationally efficient algorithms detect correlations between environmental and genetic variation while simultaneously inferring background levels of population structure. Comparing these new algorithms with related methods provides evidence that LFMM can efficiently estimate random effects due to population history and isolation-by-distance patterns when computing gene-environment correlations, and decrease the number of false-positive associations in genome scans. We then apply these models to plant and human genetic data, identifying several genes with functions related to development that exhibit strong correlations with climatic gradients.

q-bio.PE↗

arXiv2013

Fast Inference of Admixture Coefficients Using Sparse Non-negative Matrix Factorization Algorithms

Inference of individual admixture coefficients, which is important for population genetic and association studies, is commonly performed using compute-intensive likelihood algorithms. With the availability of large population genomic data sets, fast versions of likelihood algorithms have attracted considerable attention. Reducing the computational burden of estimation algorithms remains, however, a major challenge. Here, we present a fast and efficient method for estimating individual admixture coefficients based on sparse non-negative matrix factorization algorithms. We implemented our method in the computer program sNMF, and applied it to human and plant genomic data sets. The performances of sNMF were then compared to the likelihood algorithm implemented in the computer program ADMIXTURE. Without loss of accuracy, sNMF computed estimates of admixture coefficients within run-times approximately 10 to 30 times faster than those of ADMIXTURE.

q-bio.PE↗

arXiv2012

The OS* Algorithm: a Joint Approach to Exact Optimization and Sampling

Most current sampling algorithms for high-dimensional distributions are based on MCMC techniques and are approximate in the sense that they are valid only asymptotically. Rejection sampling, on the other hand, produces valid samples, but is unrealistically slow in high-dimension spaces. The OS* algorithm that we propose is a unified approach to exact optimization and sampling, based on incremental refinements of a functional upper bound, which combines ideas of adaptive rejection sampling and of A* optimization search. We show that the choice of the refinement can be done in a way that ensures tractability in high-dimension spaces, and we present first experiments in two different settings: inference in high-order HMMs and in large discrete graphical models.

cs.AI↗

arXiv2011

Linear programming problems for frontier estimation

We propose new estimates for the frontier of a set of points. They are defined as kernel estimates covering all the points and whose associated support is of smallest surface. The estimates are written as linear combinatio- ns of kernel functions applied to the points of the sample. The coefficients of the linear combination are then computed by solving a linear programming problem. In the general case, the solution of the optimizat- ion problem is sparse, that is, only a few coefficients are non zero. The corresponding points play the role of support vectors in the statistical learning theory. The L_1 error between the estimated and the true frontiers is shown to be almost surely converging to zero, and the rate of convergence is provided. The behaviour of the estimates on finite sample situations is illustrated on some simulations.

stat.ME↗