Search Results for author: Michalis Vazirgiannis

Found 80 papers, 40 papers with code

Improving Graph Neural Networks at Scale: Combining Approximate PageRank and CoreRank

1 code implementation8 Nov 2022 Ariel R. Ramos Vela, Johannes F. Lutzeyer, Anastasios Giovanidis, Michalis Vazirgiannis

Graph Neural Networks (GNNs) have achieved great successes in many learning tasks performed on graph structures.

Weisfeiler and Leman go Hyperbolic: Learning Distance Preserving Node Representations

no code implementations4 Nov 2022 Giannis Nikolentzos, Michail Chatzianastasis, Michalis Vazirgiannis

In recent years, graph neural networks (GNNs) have emerged as a promising tool for solving machine learning problems on graphs.

Graph Classification

Questioning the Validity of Summarization Datasets and Improving Their Factual Consistency

no code implementations31 Oct 2022 Yanzhu Guo, Chloé Clavel, Moussa Kamal Eddine, Michalis Vazirgiannis

Due to this lack of well-defined formulation, a large number of popular abstractive summarization datasets are constructed in a manner that neither guarantees validity nor meets one of the most essential criteria of summarization: factual consistency.

Abstractive Text Summarization

DATScore: Evaluating Translation with Data Augmented Translations

no code implementations12 Oct 2022 Moussa Kamal Eddine, Guokan Shang, Michalis Vazirgiannis

The rapid development of large pretrained language models has revolutionized not only the field of Natural Language Generation (NLG) but also its evaluation.

Data Augmentation Language Modelling +4

Word Sense Induction with Hierarchical Clustering and Mutual Information Maximization

no code implementations11 Oct 2022 Hadi Abdine, Moussa Kamal Eddine, Michalis Vazirgiannis, Davide Buscaldi

In this paper, we propose a novel unsupervised method based on hierarchical clustering and invariant information clustering (IIC).

Language Modelling Word Sense Induction

Abstractive Meeting Summarization: A Survey

1 code implementation8 Aug 2022 Virgile Rennard, Guokan Shang, Julie Hunter, Michalis Vazirgiannis

Recent advances in deep learning, and especially the invention of encoder-decoder architectures, has significantly improved the performance of abstractive summarization systems.

Abstractive Text Summarization Meeting Summarization

Time Series Forecasting Models Copy the Past: How to Mitigate

no code implementations27 Jul 2022 Chrysoula Kosma, Giannis Nikolentzos, Nancy Xu, Michalis Vazirgiannis

Recently neural network architectures have been widely applied to the problem of time series forecasting.

Time Series Forecasting

Image Keypoint Matching using Graph Neural Networks

no code implementations27 May 2022 Nancy Xu, Giannis Nikolentzos, Michalis Vazirgiannis, Henrik Boström

Image matching is a key component of many tasks in computer vision and its main objective is to find correspondences between features extracted from different natural images.

Graph Matching

Graph Ordering Attention Networks

1 code implementation11 Apr 2022 Michail Chatzianastasis, Johannes F. Lutzeyer, George Dasoulas, Michalis Vazirgiannis

The GOAT model demonstrates its increased performance in modeling graph metrics that capture complex information, such as the betweenness centrality and the effective size of a node.

Node Classification

AraBART: a Pretrained Arabic Sequence-to-Sequence Model for Abstractive Summarization

no code implementations21 Mar 2022 Moussa Kamal Eddine, Nadi Tomeh, Nizar Habash, Joseph Le Roux, Michalis Vazirgiannis

Like most natural language understanding and generation tasks, state-of-the-art models for summarization are transformer-based sequence-to-sequence architectures that are pretrained on large corpora.

Abstractive Text Summarization Natural Language Understanding

NLP Research and Resources at DaSciM, Ecole Polytechnique

no code implementations1 Dec 2021 Hadi Abdine, Yanzhu Guo, Moussa Kamal Eddine, Giannis Nikolentzos, Stamatis Outsios, Guokan Shang, Christos Xypolopoulos, 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.

Permute Me Softly: Learning Soft Permutations for Graph Representations

1 code implementation5 Oct 2021 Giannis Nikolentzos, George Dasoulas, Michalis Vazirgiannis

In this paper, we propose a new graph neural network model, so-called $\pi$-GNN which learns a "soft" permutation (i. e., doubly stochastic) matrix for each graph, and thus projects all graphs into a common vector space.

Graph Classification Graph Regression

Geometric Random Walk Graph Neural Networks via Implicit Layers

no code implementations29 Sep 2021 Giannis Nikolentzos, Michalis Vazirgiannis

The proposed model retains the transparency of Random Walk Graph Neural Networks since its first layer also consists of a number of trainable ``hidden graphs'' which are compared against the input graphs using the geometric random walk kernel.

Graph Classification

Node Feature Kernels Increase Graph Convolutional Network Robustness

1 code implementation4 Sep 2021 Mohamed El Amine Seddik, Changmin Wu, Johannes F. Lutzeyer, Michalis Vazirgiannis

The robustness of the much-used Graph Convolutional Networks (GCNs) to perturbations of their input is becoming a topic of increasing importance.

Node Classification

Sparsifying the Update Step in Graph Neural Networks

1 code implementation2 Sep 2021 Johannes F. Lutzeyer, Changmin Wu, Michalis Vazirgiannis

In this paper we conduct a structured, empirical study of the effect of sparsification on the trainable part of MPNNs known as the Update step.

Learning Graph Representations for Influence Maximization

no code implementations10 Aug 2021 George Panagopoulos, Nikolaos Tziortziotis, Fragkiskos D. Malliaros, Michalis Vazirgiannis

We develop GLIE, a Graph Neural Network (GNN) that inherently parameterizes an upper bound of influence estimation and train it on small simulated graphs.

Combinatorial Optimization

Cold Start Similar Artists Ranking with Gravity-Inspired Graph Autoencoders

1 code implementation2 Aug 2021 Guillaume Salha-Galvan, Romain Hennequin, Benjamin Chapus, Viet-Anh Tran, Michalis Vazirgiannis

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.

Link Prediction

Graph-based Neural Architecture Search with Operation Embeddings

1 code implementation11 May 2021 Michail Chatzianastasis, George Dasoulas, Georgios Siolas, Michalis Vazirgiannis

Neural Architecture Search (NAS) has recently gained increased attention, as a class of approaches that automatically searches in an input space of network architectures.

Neural Architecture Search

Evaluation Of Word Embeddings From Large-Scale French Web Content

3 code implementations5 May 2021 Hadi Abdine, Christos Xypolopoulos, Moussa Kamal Eddine, Michalis Vazirgiannis

Adding that pretrained word vectors on huge text corpus achieved high performance in many different NLP tasks.

Word Embeddings

Ego-based Entropy Measures for Structural Representations on Graphs

no code implementations17 Feb 2021 George Dasoulas, Giannis Nikolentzos, Kevin Scaman, Aladin Virmaux, Michalis Vazirgiannis

Machine learning on graph-structured data has attracted high research interest due to the emergence of Graph Neural Networks (GNNs).

Graph Classification

How COVID-19 Is Changing Our Language : Detecting Semantic Shift in Twitter Word Embeddings

no code implementations15 Feb 2021 Yanzhu Guo, Christos Xypolopoulos, Michalis Vazirgiannis

We employ an alignment-based approach to compare these embeddings with a general-purpose Twitter embedding unrelated to COVID-19.

Word Embeddings

Learning Parametrised Graph Shift Operators

1 code implementation ICLR 2021 George Dasoulas, Johannes Lutzeyer, Michalis Vazirgiannis

In many domains data is currently represented as graphs and therefore, the graph representation of this data becomes increasingly important in machine learning.

Graph Classification

An Empirical Study of the Expressiveness of Graph Kernels and Graph Neural Networks

no code implementations1 Jan 2021 Giannis Nikolentzos, George Panagopoulos, Michalis Vazirgiannis

Graph neural networks and graph kernels have achieved great success in solving machine learning problems on graphs.

Graph Similarity

Analysing the Update step in Graph Neural Networks via Sparsification

no code implementations1 Jan 2021 Changmin Wu, Johannes F. Lutzeyer, Michalis Vazirgiannis

In recent years, Message-Passing Neural Networks (MPNNs), the most prominent Graph Neural Network (GNN) framework, have celebrated much success in the analysis of graph-structured data.

Random Walk Graph Neural Networks

no code implementations NeurIPS 2020 Giannis Nikolentzos, Michalis Vazirgiannis

The first layer of the model consists of a number of trainable ``hidden graphs'' which are compared against the input graphs using a random walk kernel to produce graph representations.

Graph Classification

BARThez: a Skilled Pretrained French Sequence-to-Sequence Model

6 code implementations EMNLP 2021 Moussa Kamal Eddine, Antoine J. -P. Tixier, Michalis Vazirgiannis

We show BARThez to be very competitive with state-of-the-art BERT-based French language models such as CamemBERT and FlauBERT.

 Ranked #1 on Text Summarization on OrangeSum (using extra training data)

FLUE Natural Language Understanding +4

Transfer Graph Neural Networks for Pandemic Forecasting

4 code implementations10 Sep 2020 George Panagopoulos, Giannis Nikolentzos, Michalis Vazirgiannis

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.

Meta-Learning Representation Learning +1

Predicting conversions in display advertising based on URL embeddings

no code implementations27 Aug 2020 Yang Qiu, Nikolaos Tziortziotis, Martial Hue, Michalis Vazirgiannis

Online display advertising is growing rapidly in recent years thanks to the automation of the ad buying process.

Hcore-Init: Neural Network Initialization based on Graph Degeneracy

no code implementations16 Apr 2020 Stratis Limnios, George Dasoulas, Dimitrios M. Thilikos, Michalis Vazirgiannis

As a multipartite graph is a combination of bipartite graphs, that are in turn the incidence graphs of hypergraphs, we design k-hypercore decomposition, the hypergraph analogue of k-core degeneracy.

Graph Mining

Unsupervised Word Polysemy Quantification with Multiresolution Grids of Contextual Embeddings

1 code implementation EACL 2021 Christos Xypolopoulos, Antoine J. -P. Tixier, Michalis Vazirgiannis

A valuable by-product of our method is the ability to sample, at no extra cost, sentences containing different senses of a given word.

EvoNet: A Neural Network for Predicting the Evolution of Dynamic Graphs

no code implementations2 Mar 2020 Changmin Wu, Giannis Nikolentzos, Michalis Vazirgiannis

Then, we employ a generative model which predicts the topology of the graph at the next time step and constructs a graph instance that corresponds to that topology.

Graph Mining

Ego-based Entropy Measures for Structural Representations

no code implementations1 Mar 2020 George Dasoulas, Giannis Nikolentzos, Kevin Scaman, Aladin Virmaux, Michalis Vazirgiannis

Moreover, on graph classification tasks, we suggest the utilization of the generated structural embeddings for the transformation of an attributed graph structure into a set of augmented node attributes.

General Classification Graph Classification

FastGAE: Scalable Graph Autoencoders with Stochastic Subgraph Decoding

2 code implementations5 Feb 2020 Guillaume Salha, Romain Hennequin, Jean-Baptiste Remy, Manuel Moussallam, Michalis Vazirgiannis

Graph autoencoders (AE) and variational autoencoders (VAE) are powerful node embedding methods, but suffer from scalability issues.

Simple and Effective Graph Autoencoders with One-Hop Linear Models

1 code implementation21 Jan 2020 Guillaume Salha, Romain Hennequin, Michalis Vazirgiannis

Over the last few years, graph autoencoders (AE) and variational autoencoders (VAE) emerged as powerful node embedding methods, with promising performances on challenging tasks such as link prediction and node clustering.

Link Prediction Node Clustering

An Ensemble Method for Producing Word Representations focusing on the Greek Language

1 code implementation loresmt (AACL) 2020 Michalis Lioudakis, Stamatis Outsios, Michalis Vazirgiannis

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.

Keep It Simple: Graph Autoencoders Without Graph Convolutional Networks

1 code implementation2 Oct 2019 Guillaume Salha, Romain Hennequin, Michalis Vazirgiannis

Graph autoencoders (AE) and variational autoencoders (VAE) recently emerged as powerful node embedding methods, with promising performances on challenging tasks such as link prediction and node clustering.

Link Prediction Node Clustering

Message Passing Attention Networks for Document Understanding

2 code implementations17 Aug 2019 Giannis Nikolentzos, Antoine J. -P. Tixier, Michalis Vazirgiannis

In this paper, we represent documents as word co-occurrence networks and propose an application of the message passing framework to NLP, the Message Passing Attention network for Document understanding (MPAD).

text-classification Text Classification

Bidirectional Context-Aware Hierarchical Attention Network for Document Understanding

1 code implementation16 Aug 2019 Jean-Baptiste Remy, Antoine Jean-Pierre Tixier, Michalis Vazirgiannis

The Hierarchical Attention Network (HAN) has made great strides, but it suffers a major limitation: at level 1, each sentence is encoded in complete isolation.

Abstractive Text Summarization Topic Classification

Gravity-Inspired Graph Autoencoders for Directed Link Prediction

3 code implementations23 May 2019 Guillaume Salha, Stratis Limnios, Romain Hennequin, Viet Anh Tran, Michalis Vazirgiannis

Graph autoencoders (AE) and variational autoencoders (VAE) recently emerged as powerful node embedding methods.

Link Prediction

Graph Kernels: A Survey

no code implementations27 Apr 2019 Giannis Nikolentzos, Giannis Siglidis, Michalis Vazirgiannis

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.

Graph Classification

Energy-based Self-attentive Learning of Abstractive Communities for Spoken Language Understanding

1 code implementation Asian Chapter of the Association for Computational Linguistics 2020 Guokan Shang, Antoine Jean-Pierre Tixier, Michalis Vazirgiannis, Jean-Pierre Lorré

Abstractive community detection is an important spoken language understanding task, whose goal is to group utterances in a conversation according to whether they can be jointly summarized by a common abstractive sentence.

Community Detection Spoken Language Understanding

Multi-task Learning for Influence Estimation and Maximization

3 code implementations18 Apr 2019 George Panagopoulos, Fragkiskos D. Malliaros, Michalis Vazirgiannis

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).

Multi-Task Learning Representation Learning

Evaluation of Greek Word Embeddings

no code implementations LREC 2020 Stamatis Outsios, Christos Karatsalos, Konstantinos Skianis, Michalis Vazirgiannis

Since word embeddings have been the most popular input for many NLP tasks, evaluating their quality is of critical importance.

Word Embeddings

Randomised Bayesian Least-Squares Policy Iteration

no code implementations6 Apr 2019 Nikolaos Tziortziotis, Christos Dimitrakakis, Michalis Vazirgiannis

We introduce Bayesian least-squares policy iteration (BLSPI), an off-policy, model-free, policy iteration algorithm that uses the Bayesian least-squares temporal-difference (BLSTD) learning algorithm to evaluate policies.

Thompson Sampling

A Degeneracy Framework for Scalable Graph Autoencoders

1 code implementation23 Feb 2019 Guillaume Salha, Romain Hennequin, Viet Anh Tran, Michalis Vazirgiannis

In this paper, we present a general framework to scale graph autoencoders (AE) and graph variational autoencoders (VAE).

Message Passing Graph Kernels

1 code implementation7 Aug 2018 Giannis Nikolentzos, Michalis Vazirgiannis

The first component is a kernel between vertices, while the second component is a kernel between graphs.

Graph Similarity

Perturb and Combine to Identify Influential Spreaders in Real-World Networks

no code implementations13 Jul 2018 Antoine J. -P. Tixier, Maria-Evgenia G. Rossi, Fragkiskos D. Malliaros, Jesse Read, Michalis Vazirgiannis

Some of the most effective influential spreader detection algorithms are unstable to small perturbations of the network structure.

Orthogonal Matching Pursuit for Text Classification

1 code implementation WS 2018 Konstantinos Skianis, Nikolaos Tziortziotis, Michalis Vazirgiannis

In text classification, the problem of overfitting arises due to the high dimensionality, making regularization essential.

Classification General Classification +3

GraKeL: A Graph Kernel Library in Python

1 code implementation6 Jun 2018 Giannis Siglidis, Giannis Nikolentzos, Stratis Limnios, Christos Giatsidis, Konstantinos Skianis, Michalis Vazirgiannis

The problem of accurately measuring the similarity between graphs is at the core of many applications in a variety of disciplines.

General Classification Graph Classification

Fusing Document, Collection and Label Graph-based Representations with Word Embeddings for Text Classification

1 code implementation WS 2018 Konstantinos Skianis, Fragkiskos Malliaros, Michalis Vazirgiannis

Contrary to the traditional Bag-of-Words approach, we consider the Graph-of-Words(GoW) model in which each document is represented by a graph that encodes relationships between the different terms.

General Classification Sentiment Analysis +3

Adaptive Submodular Influence Maximization with Myopic Feedback

no code implementations23 Apr 2017 Guillaume Salha, Nikolaos Tziortziotis, Michalis Vazirgiannis

This paper examines the problem of adaptive influence maximization in social networks.

Social and Information Networks

Word Embeddings for the Construction Domain

1 code implementation28 Oct 2016 Antoine J. -P. Tixier, Michalis Vazirgiannis, Matthew R. Hallowell

Our vectors were obtained by running word2vec on an 11M-word corpus that we created from scratch by leveraging freely-accessible online sources of construction-related text.

General Classification Word Embeddings

Text Relatedness Based on a Word Thesaurus

no code implementations15 Jan 2014 George Tsatsaronis, Iraklis Varlamis, Michalis Vazirgiannis

Without doubt, a measure of relatedness between text segments must take into account both the lexical and the semantic relatedness between words.

Retrieval Sentence Similarity +2

Clustering and Community Detection in Directed Networks: A Survey

no code implementations5 Aug 2013 Fragkiskos D. Malliaros, Michalis Vazirgiannis

Networks (or graphs) appear as dominant structures in diverse domains, including sociology, biology, neuroscience and computer science.

Community Detection Graph Clustering +1

Cannot find the paper you are looking for? You can Submit a new open access paper.