arXiv subjects
Publications and source records attributed to Michalis Vazirgiannis.
DaSciM (Data Science and Mining) part of LIX at Ecole Polytechnique, established in 2013 and since then producing research results in the area of large scale data analysis via methods of machine and deep learning. The group has been specifically active in the area of NLP and text mining with interesting results at methodological and resources level. Here follow our different contributions of interest to the AFIA community.
Graph kernels have attracted a lot of attention during the last decade, and have evolved into a rapidly developing branch of learning on structured data. During the past 20 years, the considerable research activity that occurred in the field resulted in the development of dozens of graph kernels, each focusing on specific structural properties of graphs. Graph kernels have proven successful in a wide range of domains, ranging from social networks to bioinformatics. The goal of this survey is to provide a unifying view of the literature on graph kernels. In particular, we present a comprehensive overview of a wide range of graph kernels. Furthermore, we perform an experimental evaluation of several of those kernels on publicly available datasets, and provide a comparative study. Finally, we discuss key applications of graph kernels, and outline some challenges that remain to be addressed.
Fast and reliable evaluation metrics are key to R&D progress. While traditional natural language generation metrics are fast, they are not very reliable. Conversely, new metrics based on large pretrained language models are much more reliable, but require significant computational resources. In this paper, we propose FrugalScore, an approach to learn a fixed, low cost version of any expensive NLG metric, while retaining most of its original performance. Experiments with BERTScore and MoverScore on summarization and translation show that FrugalScore is on par with the original metrics (and sometimes better), while having several orders of magnitude less parameters and running several times faster. On average over all learned metrics, tasks, and variants, FrugalScore retains 96.8% of the performance, runs 24 times faster, and has 35 times less parameters than the original metrics. We make our trained metrics publicly available, to benefit the entire NLP community and in particular researchers and practitioners with limited resources.
We introduce BERTweetFR, the first large-scale pre-trained language model for French tweets. Our model is initialized using the general-domain French language model CamemBERT which follows the base architecture of RoBERTa. Experiments show that BERTweetFR outperforms all previous general-domain French language models on two downstream Twitter NLP tasks of offensiveness identification and named entity recognition. The dataset used in the offensiveness detection task is first created and annotated by our team, filling in the gap of such analytic datasets in French. We make our model publicly available in the transformers library with the aim of promoting future research in analytic tasks for French tweets.
Neural Architecture Search (NAS) has recently gained increased attention, as a class of approaches that automatically searches in an input space of network architectures. A crucial part of the NAS pipeline is the encoding of the architecture that consists of the applied computational blocks, namely the operations and the links between them. Most of the existing approaches either fail to capture the structural properties of the architectures or use hand-engineered vector to encode the operator information. In this paper, we propose the replacement of fixed operator encoding with learnable representations in the optimization process. This approach, which effectively captures the relations of different operations, leads to smoother and more accurate representations of the architectures and consequently to improved performance of the end task. Our extensive evaluation in ENAS benchmark demonstrates the effectiveness of the proposed operation embeddings to the generation of highly accurate models, achieving state-of-the-art performance. Finally, our method produces top-performing architectures that share similar operation and graph patterns, highlighting a strong correlation between the structural properties of the architecture and its performance.
On an artist's profile page, music streaming services frequently recommend a ranked list of "similar artists" that fans also liked. However, implementing such a feature is challenging for new artists, for which usage data on the service (e.g. streams or likes) is not yet available. In this paper, we model this cold start similar artists ranking problem as a link prediction task in a directed and attributed graph, connecting artists to their top-k most similar neighbors and incorporating side musical information. Then, we leverage a graph autoencoder architecture to learn node embedding representations from this graph, and to automatically rank the top-k most similar neighbors of new artists using a gravity-inspired mechanism. We empirically show the flexibility and the effectiveness of our framework, by addressing a real-world cold start similar artists ranking problem on a global music streaming service. Along with this paper, we also publicly release our source code as well as the industrial graph data from our experiments.
Graph autoencoders (AE) and variational autoencoders (VAE) are powerful node embedding methods, but suffer from scalability issues. In this paper, we introduce FastGAE, a general framework to scale graph AE and VAE to large graphs with millions of nodes and edges. Our strategy, based on an effective stochastic subgraph decoding scheme, significantly speeds up the training of graph AE and VAE while preserving or even improving performances. We demonstrate the effectiveness of FastGAE on various real-world graphs, outperforming the few existing approaches to scale graph AE and VAE by a wide margin.
The recent outbreak of COVID-19 has affected millions of individuals around the world and has posed a significant challenge to global healthcare. From the early days of the pandemic, it became clear that it is highly contagious and that human mobility contributes significantly to its spread. In this paper, we study the impact of population movement on the spread of COVID-19, and we capitalize on recent advances in the field of representation learning on graphs to capture the underlying dynamics. Specifically, we create a graph where nodes correspond to a country's regions and the edge weights denote human mobility from one region to another. Then, we employ graph neural networks to predict the number of future cases, encoding the underlying diffusion patterns that govern the spread into our learning model. Furthermore, to account for the limited amount of training data, we capitalize on the pandemic's asynchronous outbreaks across countries and use a model-agnostic meta-learning based method to transfer knowledge from one country's model to another's. We compare the proposed approach against simple baselines and more traditional forecasting techniques in 3 European countries. Experimental results demonstrate the superiority of our method, highlighting the usefulness of GNNs in epidemiological prediction. Transfer learning provides the best model, highlighting its potential to improve the accuracy of the predictions in case of secondary waves, if data from past/parallel outbreaks is utilized.
Scientific impact has been the center of extended debate regarding its accuracy and reliability. From hiring committees in academic institutions to governmental agencies that distribute funding, an author's scientific success as measured by the h-index is a vital point to their career. The objective of this work is to investigate whether the collaboration patterns of an author are good predictors of the author's future $h$-index. Although not directly related to each other, a more intense collaboration can result into increased productivity which can potentially have an impact on the author's future $h$-index. In this paper, we capitalize on recent advances in graph neural networks and we examine the possibility of predicting the $h$-index relying solely on the author's collaboration and the textual content of a subset of their papers. We perform our experiments on a large-scale network consisting of more than $1$ million authors that have published papers in computer science venues and more than $37$ million edges. The task is a six-months-ahead forecast, i.e. what the $h$-index of each author will be after six months. Our experiments indicate that there is indeed some relationship between the future $h$-index of an author and their structural role in the co-authorship network. Furthermore, we found that the proposed method outperforms standard machine learning techniques based on simple graph metrics along with node representations learned from the textual content of the author's papers.
Machine learning on graph-structured data has attracted high research interest due to the emergence of Graph Neural Networks (GNNs). Most of the proposed GNNs are based on the node homophily, i.e neighboring nodes share similar characteristics. However, in many complex networks, nodes that lie to distant parts of the graph share structurally equivalent characteristics and exhibit similar roles (e.g chemical properties of distant atoms in a molecule, type of social network users). A growing literature proposed representations that identify structurally equivalent nodes. However, most of the existing methods require high time and space complexity. In this paper, we propose VNEstruct, a simple approach, based on entropy measures of the neighborhood's topology, for generating low-dimensional structural representations, that is time-efficient and robust to graph perturbations. Empirically, we observe that VNEstruct exhibits robustness on structural role identification tasks. Moreover, VNEstruct can achieve state-of-the-art performance on graph classification, without incorporating the graph structure information in the optimization, in contrast to GNN competitors.
Words are malleable objects, influenced by events that are reflected in written texts. Situated in the global outbreak of COVID-19, our research aims at detecting semantic shifts in social media language triggered by the health crisis. With COVID-19 related big data extracted from Twitter, we train separate word embedding models for different time periods after the outbreak. We employ an alignment-based approach to compare these embeddings with a general-purpose Twitter embedding unrelated to COVID-19. We also compare our trained embeddings among them to observe diachronic evolution. Carrying out case studies on a set of words chosen by topic detection, we verify that our alignment approach is valid. Finally, we quantify the size of global semantic shift by a stability measure based on back-and-forth rotational alignment.
The number of senses of a given word, or polysemy, is a very subjective notion, which varies widely across annotators and resources. We propose a novel method to estimate polysemy, based on simple geometry in the contextual embedding space. Our approach is fully unsupervised and purely data-driven. We show through rigorous experiments that our rankings are well correlated (with strong statistical significance) with 6 different rankings derived from famous human-constructed resources such as WordNet, OntoNotes, Oxford, Wikipedia etc., for 6 different standard metrics. We also visualize and analyze the correlation between the human rankings. A valuable by-product of our method is the ability to sample, at no extra cost, sentences containing different senses of a given word. Finally, the fully unsupervised nature of our method makes it applicable to any language. Code and data are publicly available at https://github.com/ksipos/polysemy-assessment . The paper was accepted as a long paper at EACL 2021.
Inductive transfer learning has taken the entire NLP field by storm, with models such as BERT and BART setting new state of the art on countless NLU tasks. However, most of the available models and research have been conducted for English. In this work, we introduce BARThez, the first large-scale pretrained seq2seq model for French. Being based on BART, BARThez is particularly well-suited for generative tasks. We evaluate BARThez on five discriminative tasks from the FLUE benchmark and two generative tasks from a novel summarization dataset, OrangeSum, that we created for this research. We show BARThez to be very competitive with state-of-the-art BERT-based French language models such as CamemBERT and FlauBERT. We also continue the pretraining of a multilingual BART on BARThez' corpus, and show our resulting model, mBARThez, to significantly boost BARThez' generative performance. Code, data and models are publicly available.
We address the problem of influence maximization when the social network is accompanied by diffusion cascades. In prior works, such information is used to compute influence probabilities, which is utilized by stochastic diffusion models in influence maximization. Motivated by the recent criticism on the effectiveness of diffusion models as well as the galloping advancements in influence learning, we propose IMINFECTOR (Influence Maximization with INFluencer vECTORs), a unified approach that uses representations learned from diffusion cascades to perform model-independent influence maximization that scales in real-world datasets. The first part of our methodology is a multi-task neural network that learns embeddings of nodes that initiate cascades (influencer vectors) and embeddings of nodes that participate in them (susceptible vectors). The norm of an influencer vector captures the ability of the node to create lengthy cascades and is used to estimate the expected influence spread and reduce the number of candidate seeds. In addition, the combination of influencer and susceptible vectors form the diffusion probabilities between nodes. These are used to reformulate the network as a bipartite graph and propose a greedy solution to influence maximization that retains the theoretical guarantees.We a pply our method in three sizable networks with diffusion cascades and evaluate it using cascades from future time steps. IMINFECTOR is able to scale in all of them and outperforms various competitive algorithms and metrics from the diverse landscape of influence maximization in terms of efficiency and seed set quality.
In this paper we present a new ensemble method, Continuous Bag-of-Skip-grams (CBOS), that produces high-quality word representations putting emphasis on the modern Greek language. The CBOS method combines the pioneering approaches for learning word representations: Continuous Bag-of-Words (CBOW) and Continuous Skip-gram. These methods are compared through intrinsic and extrinsic evaluation tasks on three different sources of data: the English Wikipedia corpus, the modern Greek Wikipedia corpus, and the modern Greek Web Content corpus. By comparing these methods across different tasks and datasets, it is evident that the CBOS method achieves state-of-the-art performance.
Online display advertising is growing rapidly in recent years thanks to the automation of the ad buying process. Real-time bidding (RTB) allows the automated trading of ad impressions between advertisers and publishers through real-time auctions. In order to increase the effectiveness of their campaigns, advertisers should deliver ads to the users who are highly likely to be converted (i.e., purchase, registration, website visit, etc.) in the near future. In this study, we introduce and examine different models for estimating the probability of a user converting, given their history of visited URLs. Inspired by natural language processing, we introduce three URL embedding models to compute semantically meaningful URL representations. To demonstrate the effectiveness of the different proposed representation and conversion prediction models, we have conducted experiments on real logged events collected from an advertising platform.
Graph neural networks (GNNs) have emerged recently as a powerful architecture for learning node and graph representations. Standard GNNs have the same expressive power as the Weisfeiler-Leman test of graph isomorphism in terms of distinguishing non-isomorphic graphs. However, it was recently shown that this test cannot identify fundamental graph properties such as connectivity and triangle freeness. We show that GNNs also suffer from the same limitation. To address this limitation, we propose a more expressive architecture, k-hop GNNs, which updates a node's representation by aggregating information not only from its direct neighbors, but from its k-hop neighborhood. We show that the proposed architecture can identify fundamental graph properties. We evaluate the proposed architecture on standard node classification and graph classification datasets. Our experimental evaluation confirms our theoretical findings since the proposed model achieves performance better or comparable to standard GNNs and to state-of-the-art algorithms.
Artificial Intelligence techniques are already popular and important in the legal domain. We extract legal indicators from judicial judgment to decrease the asymmetry of information of the legal system and the access-to-justice gap. We use NLP methods to extract interesting entities/data from judgments to construct networks of lawyers and judgments. We propose metrics to rank lawyers based on their experience, wins/loss ratio and their importance in the network of lawyers. We also perform community detection in the network of judgments and propose metrics to represent the difficulty of cases capitalising on communities features.