The project ACADEMICS (mAChine LeArning & Data sciEnce for coMplex and dynamICal modelS) will combine Machine Learning (ML) and Data Science (DS) for the purpose of scientific research into two challenging directions: (a) Computing and information processing – develop new theoretical frameworks and learning algo- rithms adapted to difficult scientific contexts involving heterogeneous, irregular, error-prone, dynamic and complex data, while taking into account prior knowledge whenever it is relevant; and (b) Complex and dynamic models learning – leverage the synergy between ML and DS to devise data- driven models in two scientific domains: climate modeling, and quantitative understanding of social systems. Focusing on these two case studies, the project will tackle the key issue of how to learn intricate models from numerous, heterogeneous and dynamic data.
The project is funded by IDEX Lyon for the amount of 1.2M euro for the period 2018-2021. Participant laboratories are LP and LIP from ENS Lyon, LIRIS from Université Lyon 1, and LabHC from the Université Jean Monnet.
The aim of the DyLNet project (Language Dynamics, Linguistic Learning, and Sociability at Preschool) is to observe and characterize the relations between child socialization and oral language learning during the preschool period by means of an innovative multidisciplinary approach that combines work in the fields of language acquisition, sociolinguistics and network science. The project implemented as large scale socio-linguistic experiment to follow ≈150 children and teaching staff at a socially mixed pre-school. The physical proximity and verbal interactions are recorded between each participants using RFID sensor technology, which captures inter-individual proximity in every 5 second and verbal interactions continuously. The experiment is run one week every month for a period of 3 years. The task, in particular, will be to examine the influence of the children’s social relations on their language development and, equally, the influence of language on these social relations.
The DyLNet project is financed by ANR for 650K euro (16-CE28-0013) for the period of 2016-2020. Participating groups are Lidilem – University of Grenoble Alpes, DANTE – ENS Lyon/Inria, LSE – University of Grenoble Alpes, Ethos – University Rennes 1, and the LLL – University pf Orleans.
MOTif (Stic AmSud)
The general goal of the MOTIf project (Mobile phone sensing of human dynamics in techno-social environment) is to understand, model, and predict individual behavior embedded in social and technological environments. We aim to understand spatiotemporal patterns of service usage of individuals to learn when, where, and what people are doing and to understand the fine-grained sociodemographic structure of society and see how the demographic characteristics of individuals in a social network correlate with the dynamics of their egocentric and global network evolution.
The MOTif project is funded by Stic AmSud by ~81K euro for the period 2018-2020. This is collaborative project to foster collaborations between France and Latin American countries. Participating groups are DANTE – ENS Lyon/Inria (France), Grandata (USA-Argentina), and the Universidade Federal de Minas Gerais, LNCC, and Pontifícia Universidade Católica de Minas Gerais (Brazil).
The goal of SoSweet is to provide a detailed understanding of the dynamic links between individuals, social structure, and language variation and change through the study of synchronic variation and diachronic evolution of the variety of French language observed on Twitter. Within the project we collect and analyse a corpus of 600 million tweets combined with the social network of the 5 million users, complemented by socio-demographic data. The SoSweet project adopts a strong interdisciplinary position, at the crossing of social media linguistics, sociolinguistics, natural language processing (NLP) and network science.
The SoSweet project is funded by ANR (15- CE38-0011) for 635K euro for the period of 2015-2019. The consortium of the project is formed by the teams of ICAR and DANTE from ENS Lyon/Inria, ALMAnaCH – Inria Paris, and Lidilem – Université Grenoble Alpes.
The purpose of the HOTNet (Higher-order representation of temporal networks) project is to develop a pipeline for the embedding of temporal networks that captures higher order correlations relevant for dynamical processes. We propose to detach from the straightforward representations of networks — as successions of static networks — by focusing on representations that better reflects the higher-order neighbourhood and temporal paths. To project plans to develop a framework that learns from this representation an embedding sufficient to estimate the outcome of spreading processes that might take place on top of the original network.
This is a small-scale collaborative project funded by the IXXI Complex System Institute to foster collaborations between MK and Laetitia Gauvin (ISI Torino) for the period of 2019-2021.
LIAISON is an exploratory project aims to develop unsupervised deep learning approaches to infer correlations/patterns that exist between dynamic linguistic variables, the mesoscopic and dynamic structure of social networks, and their socio-economic attributes. This interdisciplinary project is positioned at the crossroads of Automatic Language Processing, Network Science, Data Science and Machine Learning.
The LIAISON project was funded by Inria within the PRE framework by 50K euro for the period of 2017-2018.
Thematic Month on Networks and Learning (Labex)
This project intended to cover both the basics of and recent advances in Network Science by organizing a series of workshops and bring together world-known experts from the fields of mathematics, physics, signal processing, computer science, social science, epidemiology and linguistic to discuss and enhance our understanding about the interaction between the structure, evolution, and coupled dynamical processes of complex networks.
The project was funded by Labex MILYON in 2016 by 90K euros.