7.1.1 Pl@ntNet

• Keywords:
  Plant identification, Deep learning, Citizen science
• Functional Description:
  Pl@ntNet is a participatory platform and information system dedicated to the production of botanical data through deep learning-based plant identification. It includes 3 main front-ends, an Android app (the most advanced and the most used one), an iOs app (being currently re-developed) and a web version. The main feature of the application is to return the ranked list of the most likely species providing an image or an image set of an individual plant. In addition, Pl@ntNet’s search engine returns the images of the dataset that are the most similar to the queried observation allowing interactive validation by the users. The back-office running on the server side of the platform is based on Snoop visual search engine (a software developed by ZENITH) and on NewSQL technologies for the data management. The application is distributed in more than 200 countries (30M downloads) and allows identifying about 35K plant species at present time. The platform integrates an open access REST API (my.plantnet.org) that currently accounts for 6500 developers accounts.
• Publications:
  hal-01629195, hal-02937618, hal-03343235, hal-01182775
• Contact:
  Alexis Joly
• Participants:
  Antoine Affouard, Jean-Christophe Lombardo, Pierre Bonnet, Hervé Goëau, Mathias Chouet, Hugo Gresse, Julien Champ, Alexis Joly

7.1.2 ThePlantGame

• Keyword:
  Crowd-sourcing
• Functional Description:
  ThePlantGame is a participatory game whose purpose is the production of big taxonomic data to improve our knowledge of biodiversity. One major contribution is the active training of the users based on innovative sub-task creation and assignment processes that are adaptive to the increasing skills of the user. Thousands of players are registered and produce on average about tens new validated plant observations per day. The accuracy of the produced taxonnomic tags is very high (about 95%), which is quite impressive considering the fact that a majority of users are beginners when they start playing.
• Publication:
  hal-01629149
• Contact:
  Alexis Joly
• Participants:
  Maximilien Servajean, Alexis Joly

7.1.3 Savime

• Name:
  Simulation And Visualization IN-Memory
• Keywords:
  Data management., Distributed Data Management
• Functional Description:
  SAVIME is a multi-dimensional array DBMS for scientific applications. It supports a novel data model called TARS (Typed ARray Schema), which extends the basic array data model with typed arrays. In TARS, the support of application dependent data characteristics is provided through the definition of TAR objects, ready to be manipulated by TAR operators. This approach provides much flexibility for capturing internal data layouts through mapping functions, which makes data ingestion independent of how simulation data has been produced, thus minimizing ingestion time.
• Publication:
  lirmm-01620376
• Contact:
  Patrick Valduriez
• Participants:
  Hermano Lustosa, Fabio Porto, Patrick Valduriez
• Partner:
  LNCC - Laboratório Nacional de Computação Científica

7.1.4 OpenAlea

• Keywords:
  Bioinformatics, Biology
• Functional Description:
  OpenAlea is an open source project primarily aimed at the plant research community. It is a distributed collaborative effort to develop Python libraries and tools that address the needs of current and future works in Plant Architecture modeling. It includes modules to analyze, visualize and model the functioning and growth of plant architecture. It was formally developed in the Inria VirtualPlants team.
• Release Contributions:
  OpenAlea 2.0 adds to OpenAlea 1.0 a high-level formalism dedicated to the modeling of morphogenesis that makes it possible to use several modeling paradigms (Blackboard, L-systems, Agents, Branching processes, Cellular Automata) expressed with different languages (Python, L-Py, R, Visual Programming, ...) to analyse and simulate shapes and their development.
• Publications:
  hal-01166298, hal-00831811
• Contact:
  Christophe Pradal
• Participants:
  Christian Fournier, Christophe Godin, Christophe Pradal, Frédéric Boudon, Patrick Valduriez, Esther Pacitti, Yann Guédon
• Partners:
  CIRAD, INRAE

7.1.5 Imitates

• Name:
  Indexing and mining Massive Time Series
• Keywords:
  Time Series, Indexing, Nearest Neighbors
• Functional Description:
  Time series indexing is at the center of many scientific works or business needs. The number and size of the series may well explode depending on the concerned domain. These data are still very difficult to handle and, often, a necessary step to handling them is in their indexing. Imitates is a Spark Library that implements two algorithms developed by Zenith. Both algorithms allow indexing massive amounts of time series (billions of series, several terabytes of data).
• Publication:
  lirmm-01886794
• Contact:
  Florent Masseglia
• Partners:
  New York University, Université Paris-Descartes

Pl@ntNet-CrowdSWE

• Contributors:
  Tanguy Lefort,Antoine AFFOUARD, Benjamin Charlier, Jean-Christophe Lombardo, Mathias Chouet, Hervé Goëau, Joseph Salmon, Pierre BONNET, Alexis Joly
• Description:
  This open dataset is a subset of the full Pl@ntNet database from the extraction of the Pl@ntNet South Western Europe (SWE) crowdsourced annotations. It contains all species identification and user votes for observations made between 2017 and 2023 in the SWE flora. In total, more than 6 699 593 plant observations are labeled by 823 251 users between January 2017 and October 2023. In addition, 98 experts were selected to obtain ground truth values for 26 811 observations. The goal of this dataset is to provide a new large scale benchmark dataset for label aggregation in a setting of crowdsourcing for image classification. It has already been used to demonstrate the current performance of the Pl@ntNet label aggregation strategy in 58.
• Dataset PID (DOI,...):
  https://­doi.­org/­10.­5281/­zenodo.­10782465
• Project link:
  https://­zenodo.­org/­records/­10782465
• Publications:
  https://­hal.­science/­hal-04603038
• Contact:
  tanguy.lefort@inria.fr
• Release contributions:
  New release, updated for metadata modifications related to anonymity (required during the reviewing session for the related publications) and founding

Seatizen Atlas

• Contributors:
  Matteo Contini, Julien Barde, Sylvain Bonhommeau, Victor Illien, Alexis Joly
• Description:

  The Seatizen Atlas dataset is a collection of geospatial and metadata-enriched images, combining participatory science data with data collected from advanced scientific platforms. This dataset integrates imagery captured using different platforms, including kitesurfs, paddleboards, snorkeling masks, Autonomous Surface Vehicles (ASVs), and Unmanned Aerial Vehicles (UAVs).

  Covering vast regions in the Southwest Indian Ocean, including Réunion Island, Seychelles, and Mauritius, the dataset comprises approximately 1,626,830 images. The images are enriched with detailed metadata, including GPS coordinates, camera settings (EXIF/XMP standards), and AI-based predictions. Additionally, 14,492 images have been manually annotated following Global Coral Reef Monitoring Network (GCRMN) and Global Biodiversity Information Facility (GBIF) standards.

  This dataset supports diverse scientific and ecological applications, such as:

  • Generating historical records of marine ecosystems with precise geolocation for monitoring ecosystem changes.
  • Producing high-resolution species distribution maps using AI-based inference models.
  • Training and evaluating computer vision algorithms on annotated subsets.
  • Creating aerial and underwater photogrammetric reconstructions using accurately georeferenced imagery.
  • Validating remote sensing models with ground-truth geolocated data.
  • Promoting FAIR (Findable, Accessible, Interoperable, Reusable) data practices through adherence to established standards (e.g., Darwin Core, GCRMN, EXIF/XMP, and COCO).

  The data workflow, including image processing, AI-based predictions, and publishing methodologies, is openly shared and reproducible through the GitHub repository: https://­github.­com/­SeatizenDOI.

• Dataset PID (DOI,...):
  10.5281/zenodo.13951435
• Project link:
  https://­seatizenmonitoring.­ifremer.­re
• Publications:
  Seatizen atlas: a collaborative dataset of underwater and aerial marine imagery. Soon available at https://­doi.­org/­10.­1038/­s41597-024-04267-z.
• Contact:
  seatizen.ifremer@gmail.com, sylvain.bonhommeau@ifremer.fr, alexis.joly@inria.fr
• Release contributions:
  • seatizen_atlas_db.gpkg: Geopackage for spatial data management.
  • session_doi.csv: List of published sessions.
  • metadata_images.csv: Metadata for published images.
  • metadata_multilabel_predictions.csv: Predictions from the AI model.
  • metadata_multilabel_annotation.csv: Annotated subset of images.
  • seatizen_atlas.qgz: QGIS project file.
  • darwincore_multilabel_annotation.zip: Darwin Core Archive with image annotations.

8.1.1 Data-Driven Model Selection for Spatio-Temporal Prediction

Participants: Fabio Porto, Patrick Valduriez.

Spatio-temporal predictive queries encompass a spatio temporal constraint, which defines a region and a target variable, and an evaluation metric. They produce the future values for the target variable computed by predictive models at each point of the spatio-temporal region. However, training temporal models at each spatial domain point can be prohibitive. In 55, we propose a data-driven approach for selecting pre-trained temporal models to be applied at each query point, which avoids training a different model for each domain point, thus saving model training time. We propose a technique to decide on the best-trained model to be applied to a point for prediction. The experimental evaluation on a case study in temperature forecasting using historical data and auto-regressive models shows that, compared to the baseline, the approach achieves equivalent predictive performance at a fraction of the total computational cost.

8.1.2 Executing Scientific Workflows with Privacy Constraints

Participants: Esther Pacitti.

Containerized and cloud environments are ideal for running scientific workflows because they offer a flexible and easily instantiated setting. Data dispersion and encryption can be adopted in this context, but not independently of workflow scaling, as this can increase the total execution time or the associated financial cost. In 49, we introduce Okinawa, a heuristic for executing workflows in containerized environments with confidentiality constraints. In 50, we introduce CYCLOPS, an approach that aims to execute workflows in cloud computing environments efficiently while considering the confidentiality constraints of the produced data and the workflow structure.

8.1.3 Federated Learning

Participants: Patrick Valduriez.

Federated Learning (FL) enables collaborative model training on distributed data over edge devices. We contributed to FL in two major ways. The first one 42 is the Asynchronous Efficient Decentralized FL framework (AEDFL) for heterogeneous devices. Extensive experimentation on four public datasets and four models demonstrates the strength of AEDFL in terms of accuracy (up to 16.3% higher), efficiency (up to 92.9% faster), and computation costs (up to 42.3% lower). The second contribution 43 is a Fisher information-based efficient curriculum federated learning framework (FibecFed) to fine-tune Large Language Models (LLMs), which comes with two novel methods, i.e., adaptive federated curriculum learning and efficient sparse parameter update. Extensive experimental results based on 10 datasets demonstrate that FibecFed yields excellent performance (up to 45.35% in terms of accuracy) and superb fine-tuning speed (up to 98.61% faster) compared with 17 baseline approaches.

8.1.4 Optimal Checkpointing for Heterogeneous Chains: How to Train Deep Neural Networks with Limited Memory

Participants: Alena Shilova, Alexis Joly.

This work was done in collaboration with the HiePacs Inria team.

Training in Feed Forward Deep Neural Networks is a memory-intensive operation which is usually performed on GPUs with limited memory capacities. This may force data scientists to limit the depth of the models or the resolution of the input data if data does not fit in the GPU memory. The re-materialization technique, whose idea comes from the checkpointing strategies developed in the Automatic Differentiation literature, allows data scientists to limit the memory requirements related to the storage of intermediate data (activations), at the cost of an increase in the computational cost. In 15, we introduce a new strategy of re-materialization of activations that significantly reduces memory usage. It consists in selecting which activations are saved and which activations are deleted during the forward phase, and then recomputing the deleted activations when they are needed during the backward phase. We propose an original computation model that combines two types of activation savings: either only storing the layer inputs, or recording the complete history of operations that produced the outputs. We prove that finding the optimal solution is NP-hard and that classical techniques from Automatic Differentiation literature do not apply. Thus, we propose a weak memory persistence property and provide a Dynamic Program to compute the optimal sequence of computations. Through extensive experiments, we show that our implementation consistently outperforms existing re-materialization approaches for a large class of networks, image sizes and batch sizes.

8.2.1 Event Detection in Time Series

Participants: Esther Pacitti, Fabio Porto, Rebecca Salles.

Event detection in times series is a basic function in surveillance and monitoring systems and has been extensively explored over the years. The book 54 provides a general taxonomy for event detection according to the specific event types: anomaly detection, change-point, and motif discovery. It discusses state-of-the-art metric evaluations for event detection methods and on online event detection, including the challenges of incremental and adaptive learning.

We also proposed a time series anomaly detection method 44, based on the Fast Fourier Transform (FFT). The method removes low-frequency components, such as trends and seasonality, which represent the normal behavior of the series, while preserving high-frequency components associated with anomalies. The experimental results show the effectiveness of the method in anomaly detection using high-pass FFT filters that have a cutoff frequency adjusted by change points.

Finally, we addressed the problem of matching detections to events in time series 48, by structuring the association problem using graph theory, as a bipartite graph matching problem and using the Hungarian algorithm as solution. The results demonstrate the effectiveness of the proposed approach, highlighting the impact of improvements in the associations between detections and events.

8.2.2 Metrics for Evaluating Event Detection in Time Series

Participants: Reza Akbarinia, Florent Masseglia, Esther Pacitti, Fabio Porto, Rebecca Salles.

Time series event detection methods are evaluated mainly by standard classification metrics that focus solely on detection accuracy. However, inaccuracy in detecting an event can often result from its preceding or delayed effects reflected in neighboring detections. To address this problem, we proposed SoftED metrics 29, a new set of metrics designed for soft evaluating event detection methods, which enable the evaluation of both detection accuracy and the degree to which their detections represent events. They improve event detection evaluation by associating events and their representative detections, incorporating temporal tolerance in over 36% of experiments compared to the usual classification metrics.

8.2.3 Subset Models for Multivariate Time Series Forecast

Participants: Reza Akbarinia, Raphael De Freitas Saldanha, Patrick Valduriez.

Multivariate time series find extensive applications in conjunction with machine learning methodologies for scenario forecasting across various domains. Nevertheless, certain domains exhibit inherent complexities and diversities, which detrimentally impact the predictive efficacy of global models. In 34, we introduce a modeling framework that accommodates shared feature characteristics and regional variations across diverse units, offering cost-effective training and robust prediction capabilities. The approach involves the following steps: (1) identifying subsets within the dataset that exhibit similar covariate patterns and training models specific to each subset; (2) mapping incoming samples to the appropriate subset based on the similarity between data distributions; and (3) using the model trained on the subset data for prediction.

8.2.4 One Health Data Analytics

Participants: Benoit Lange, Reza Akbarinia, Florent Masseglia, Christophe Pradal.

Many scientific applications require a "One Health" approach to data analytics that integrates data, models and tools from different application domains. We used this approach in two applications: antimicrobial resistance and reduction of pesticide usage.

Antimicrobial resistance (AMR) presents critical health risks for humans, animals, and the environment. In 39, we introduce the PROMISE platform, for managing and analyzing One Health data. It supports 25 academic networks and 42 partners, resolving data heterogeneity through built-in interoperability, allowing users to retain their preferred taxonomies. The platform facilitates cross-disciplinary insights into AMR, showcasing its capability to streamline and unify diverse datasets for effective analysis.

Reducing pesticide use in agricultural areas requires joint consideration of the ecological, economic and social components that contribute to this goal. This requires a "One Health" approach, integrating data, models and simulation tools at various scales, from the plant to the farm to the territory. In 27, we introduce the TRAVERSEES project, for designing a socio-ecosystem model used to simulate agricultural practice trajectories across territories, which subsequently serve as a tool for prospective analysis with stakeholders in the Barrois region (Grand-Est). The project employs a range of methodologies and engages in a transdisciplinary partnership. This approach leads to the emergence of multiple, innovative proposals for territorial transition levers and highlighted the diverse factors considered by farmers, along with varying degrees of sensitivity to these influences.

8.3.1 AI-Based Species Distribution Modeling and Mapping

Participants: Christophe Botella, Benjamin Deneu, Joaquim Estopinan, Alexis Joly, Théo Larcher, César Leblanc, François Munoz, Lukáš Picek.

Although increasing threats on biodiversity are now widely recognized, there are no accurate global maps showing whether and where species assemblages are at risk. In 21, we introduce a new Deep Species Distribution Model trained on 1M occurrences of 14K orchid species to predict their assemblages at global scale and at kilometre resolution. We propose two main indicators of the conservation status of the assemblages: (i) the proportion of threatened species, and (ii) the status of the most threatened species in the assemblage. We show and analyze the variation of these indicators at World scale and in relation to currently protected areas in Sumatra island. Global and interactive maps available online show the indicators of conservation status of orchid assemblages, with sharp spatial variations at all scales. In 45 (Neurips 2024 Dataset and Benchmarks), we designed and developed a new European-scale dataset and benchmark for Multimodal SDMs at high spatial resolution (10–50m), including more than 10k species (i.e., most of the European flora). It comprises 5M heterogeneous Presence-Only records and 90k exhaustive Presence-Absence survey records, all accompanied by diverse environmental rasters (e.g., elevation, human footprint, and soil), Sentinel-2 RGB and NIR satellite images with 10 m resolution, 20-year time series of climatic variables, and satellite time series from the Landsat program. In addition to the data, we provide an openly accessible benchmark (hosted on Kaggle), which has already attracted an active community and a set of strong baselines for single predictor/modality and multimodal approaches. All resources, e.g., the dataset, pre-trained models, and baseline methods (in the form of notebooks), are available on Kaggle, allowing one to start with the dataset literally with two mouse clicks. In 26, we propose a new method that leverages convolutional neural networks (CNNs) to capture spatial features of environmental variables in the open ocean. We considered 38 taxa comprising pelagic fishes, elasmobranchs, marine mammals, marine turtles and birds. We trained a model to predict probabilities from the environmental conditions at any specific point in space and time, using species occurrence data from the Global Biodiversity Information Facility (GBIF) and environmental data from various sources. These variables included sea surface temperature, chlorophyll concentration, salinity and fifteen others. The classifier accurately predicted the observed taxon as the most likely in 69% of cases and included the observed taxon among the top three most likely predictions in 89% of cases. Additionally, this purely correlative model was then analyzed with explicability tools to understand which variables had an influence on the model’s predictions.

8.3.2 Species-to-Habitats Classification

Participants: César Leblanc, Alexis Joly.

In 22, we worked on a deep‐learning framework for enhancing habitat identification based on species composition. We leveraged deep-learning techniques, such as transformers and tested different network architectures, feature encodings, hyperparameter tuning and noise addition strategies to identify the optimal model. Our best algorithm, applied to European habitat types, significantly improved habitat classification accuracy, achieving a more than twofold improvement compared to the previous state-of-the-art (SOTA) method on an external data set, clearly outperforming expert systems. The accuracy score on a database containing hundreds of thousands of standardized presence/absence European surveys reaches 88.74%, as assessed by expert judgment. Finally, our results showcase that species dominance is a strong marker of ecosystems and that the exact cover abundance of the flora is not required to train neural networks with predictive performances. The framework we developed can be used by researchers and practitioners to accurately classify habitats. This work was co-authored with 25 other research labs who provided data (in the context of the European Vegetation Archive), developed the initial expert system and contributed ideas and writing improvements.

8.3.3 Unseen Plant Disease Recognition

Participants: Alexis Joly.

This work was done in the context of a collaboration with Swinburne University of Technology.

Automatic plant disease recognition has a great potential for improving agricultural productivity and sustainability by providing timely, accurate, and scalable solutions for managing plant health. One of the key challenge, however, is the impracticality of collecting samples for all diseases and every plant species. In 18 and 40, we worked on the recognition of unseen plant diseases. We propose new models capable of disantangling the pathogen's characteristic information from that of the host plant, so that it can be recombined for any plant/disease pair. In 18, we introduce a Cross Learning Vision Transformer (CL-ViT) model, incorporating self-supervised learning, in contrast to the previous state-of-the-art, FF-ViT, which emphasizes conceptual feature disentanglement with a synthetic feature generation framework. In 40, we introduce a novel approach that incorporates textual data to guide the vision model in learning the representation of multiple plants and diseases.

8.3.4 Evaluation of Species Identification and Prediction Algorithms

Participants: Alexis Joly, Lukáš Picek, Hervé Goëau, Christophe Botella, Benjamin Deneu, Diego Marcos, Joaquim Estopinan, César Leblanc, Théo Larcher.

We ran a new edition of the LifeCLEF evaluation campaign 36, 37 with the involvement of hundreds of data scientists and research teams worldwide. It delivers a unique view of state-of-the-art performance on species identification and prediction problems, thanks to realistic datasets and controlled evaluation methodologies. One of the main outcome was that domain shift problems remain a major problem for the emergence of new techniques, such as passive acoustic sensors, HD images of plant cover, or remote sensing monitoring. The lack of annotated data for these new domains considerably hinders the progress of supervised methods, and alternative cross-domain methods are struggling to emerge. A great hope may lie in the use of unlabeled data, which will become increasingly available and whose use for domain adaptation or self-supervised learning is beginning to emerge as an effective solution (notably in BirdCLEF 38 and GeoLifeCLEF 46). Another very promising prospect is multi-modal model learning, which was the key to the success of the best methods for the GeoLifeCLEF challenge and has enabled improvements in other tasks, including FungiCLEF 47 and PlantCLEF 35. As far as model architectures are concerned, there is a wide disparity between the use of large-scale foundation models such as DinoV2 in PlantCLEF, SnakeCLEF, and FungiCLEF and a certain trend towards frugal architectures in GeoLifeCLEF, FungiCLEF, SnakeCLEF, and BirdCLEF. Finally, it is important to note the strength of collaborative work in the progression of the challenges. The sharing of knowledge, models, or codes, whether by the organizers or the participants themselves, has a direct impact on their subsequent developments and promotes co-construction rather than sole competition.

8.3.5 New Features in the Pl@ntNet Platform

Participants: Alexis Joly, Benjamin Deneu, Jean-Christophe Lombardo, Antoine Affouard.

Pl@ntnet is a citizen observatory that relies on AI technologies to help people identify plants with their smartphones. Over the past few years, Pl@ntNet has become one of the largest plant biodiversity observatories in the world with several million contributors. A set of new features were developed in 2024.

First, the web front-end of the application has been deeply refactorized with strong UI and UX improvements (thanks to the recruitment of an engineer expert in this domain). Most views of the application were revisited, necessitating substantial development work, including on the back-end, as numerous new data access functionalities had to be developed, impacting all layers (database, data access layer, API). Pl@ntNet web's front-end is used by about 10M users per year.

We also continued the development of new tools for managing and analyzing vegetation plot images (e.g. drone images, quadrat images, road-side views, etc.). We first optimized our tiling approach for the analysis of such HD pictures (in pytorch). This allowed to divide the computation cost by a factor 5. We also enriched the API to make this service more easily usable by external partners and we finalized the development of the web front-end thanks to the feedbacks of beta testers along the year. Those tools are already used by several projects/organizations including: (i) a Biodiversa+ pilot project on the monitoring of invasive alien species (coordinated by the Danish Ministry of the Environment through Aarhus University), (ii) two international consortia who participated to the X-prize rain forest challenge (who obtained the second and third place over 300 participants), (iii) the French project Pl@ntAgroEco.

A third key feature developed in 2024 is the possibility to manage and identifiy other plant attributes than just the species through a mechanism called "tags" (key-valued). In particular, it enables the identification and management of two key concepts for agro-ecology: (i) plant damages (pathogens, diseases, fungies, eyc.) and (ii), plant varieties (sub-specific levels in the taxonomy such as crop varities, horticicultural varieties). This required in-depth work on the database and the data access layer. The integration of these new features into the web and mobile front-ends is still under development.

A last new feature developed as part of the GUARDEN EU project is to enable the import of observations by batch within the database. Indeed, many institutions and private individuals have large batches of plant observations that they would like to share on the platform. A new interface has therefore been developed in the Pl@ntNet web application.

HPDaSc

• Title:
  High Performance Data Science
• Duration:
  2020 ->
• Coordinator:
  Fabio Porto (fporto@lncc.br)
• Partners:
  • Laboratório Nacional de Computação Científica Petrópolis (Brésil)
• Inria contact:
  Patrick Valduriez
• Summary:

  Data-intensive science refers to modern science, such as astronomy, geoscience or life science, where researchers need to manipulate and explore massive datasets produced by observation or simulation. It requires the integration of two fairly different paradigms: high-performance computing (HPC) and data science. We address the following requirements for high-performance data science (HPDaSc): Support realtime analytics and visualization (in either in situ or in transit architectures) to help make high-impact online decisions; Combine ML with analytics and simulation, which implies dealing with uncertain training data, autonomously built ML models and combine ML models and simulation models; Support scientific workflows that combine analytics, modeling and simulation, and exploit provenance in realtime and HIL (Human in the Loop) for efficient workflow execution.

  To address these requirements, we will exploit new distributed and parallel architectures and design new techniques for ML, realtime analytics and scientific workflow management. The architectures will be in the context of multisite cloud, with heterogeneous data centers with data nodes, compute nodes and GPUs. We will validate our techniques with major software systems on real applications with real data. The main systems will be OpenAlea and Pl@ntnet from Zenith and DfAnalyzer and SAVIME from the Brazilian side. The main applications will be in agronomy and plant phenotyping (with plant biologists from CIRAD and INRA), biodiversity informatics (with biodiversity scientists from LNCC and botanists from CIRAD), and oil & gas (with geoscientists from UFRJ and Petrobras).

9.2.1 Visits of international scientists

9.3.1 Horizon Europe

Pl@ntAgroEco (PEPR Agroécologie et Numérique), (2023-2027), 1.6 Meuro.

Participants: Antoine Affouard, Christophe Botella, Hervé Goëau, Hugo Gresse, Alexis Joly, Thomas Paillot.

Agroecology necessarily involves crop diversification, but also the early detection of diseases, deficiencies and stresses (hydric, etc.), as well as better management of biodiversity. The main stumbling block is that this paradigm shift in agricultural practices requires expert skills in botany, plant pathology and ecology that are not generally available to those working in the field, such as farmers or agri-food technicians. Digital technologies, and artificial intelligence in particular, can play a crucial role in removing this barrier to access to knowledge.

The aim of the Pl@ntAgroEco project will be to design, experiment with and develop new high-impact agro-ecology services within the Pl@ntNet platform. This includes : AI and plant science research ; agile development of new components within the platform; organizing participatory science programs and animating the Pl@ntNet user community. The project is leaded by ZENITH (Alexis Joly).

FishPredict (ANR), (2022-2025), 500 Keuro.

Participants: Benjamin Bourel, Alexis Joly, Maximilien Servajean, Julien Thomazo.

Fish-predict ANR project funded in the context of the IA-Biodiv challenge. The projects aims at predicting the biodiversity of reef fishes using AI technologies. Alexis Joly is co-leading of the whole project jointly with David Mouillot, marine ecologist at the MARBEC lab.

ANR PerfAnalytics (2021-2024), 100 Keuro.

Participants: Reza Akbarinia, Florent Masseglia.

The objective of the PerfAnalytics project is to analyze sport videos in order to quantify the sport performance indicators and provide feedback to coaches and athletes, particularly to French sport federations in the perspective of the Paris 2024 Olympic games. A key aspect of the project is to couple the existing technical results on human pose estimation from video with scientific methodologies from biomechanics for advanced gesture objectivation. The motion analysis from video represents a great potential for any monitoring of physical activity. In that sense, it is expected that exploitation of results will be able to address not only sport, but also the medical field for orthopedics and rehabilitation.

PPR Antibiorésistance: structuring tool "PROMISE" (2021-2024), 240 Keuro.

Participants: Reza Akbarinia, Florent Masseglia.

The objective of the PROMISE (PROfessional coMmunIty network on antimicrobial reSistancE) project is to build a large data warehouse for managing and analyzing antimicrobial resitance (AMR) data. It gathers 21 existing professional networks and 42 academic partners from three sectors, human, animal, and environment. The project is based on the following transdisciplinary and cross-sectoral pillars: i) fostering synergies to improve the one-health surveillance of antibiotic consumption and AMR, ii) data sharing for improving the knowledge of professionals, iii) improving clinical research by analyzing the shared data.

PNR "Beerisk" (2022-2025). 200K Keuro.

Participants: Reza Akbarinia, Florent Masseglia.

The objective of this project is to analyze honeybee daily mortality rates, represented as time series, in order to detect anomalies and study the lethal effects of bees exposure to pesticides.

Plan national Ecoantibio "INTERSECTION" (2024-2028), 175 Keuros

Participants: Reza Akbarinia, Florent Masseglia.

The objective of the INTERSECTION project is to produce intersectoral and territorial indicators for monitoring resistance and use of antibiotics in France, and to facilitate the use and analysis of these indicators, in a One health approach.

PEPR agroécologie et numérique "RootSystemTracker" (2024-2027), 144 Keuros

Participants: Reza Akbarinia, Christophe Pradal.

Roots play a crucial role in nutrient and water uptake, atmospheric carbon fixation, and soil interactions, significantly influencing resource use efficiency and crop resilience to environmental stresses. The objective of the RootSystemTracker project is to develop efficient methods for the spatio-temporal phenotyping of plant root architectures using heterogeneous data. This involves automatically capturing their topology and geometry over time, despite challenges such as root occlusions and variability in observation conditions.

9.4.1 Others

Regional project "DACLIM" (2023-2026), 70 Keuros

Participants: Reza Akbarinia, Florent Masseglia.

The objective of this project is to develop scalable techniques based on massive data distribution to enable the efficient detection of anomalies in large climate databases. The detection of anomalies in climate data can provide climatologists with insights into the behavior of various climatological variables, understanding of extreme events such as heatwaves and cold snaps, as well as the prediction of these types of events.

CESE consultation on the impact of AI on the environment

: The CESE (Conseil Economique, Social et Envinonemental) is one of the 3 assemblies of the French constitution, made up of elected representatives of civil society (unions, associations, companies, students, etc.). Its role is to provide advice on economic, social and environmental policies to guide public decision-making (governmental in particular). Alexis Joly took part in the consultation entitled “Impacts of artificial intelligence: risks and opportunities for the environment”. He was consulted and interviewed on several occasions and was one of the 3 experts invited to the final plenary session that voted on the recommendations.

OCDE report the advancement of the productivity of science with citizen science and artificial

OCDE report on the advancement of the productivity of science with citizen science and artificial)

10.1.1 Scientific events: organisation

10.1.2 Journal

10.1.3 Invited talks

• P. Valduriez:
  • "Data Science and Innovation"
  • IMPA, Rio de Janeiro, May 5, 2024
  • IRIT, Toulouse, July 12, 2024
  • LNCC, Petropolis, Rio de Janeiro, July 22, 2024
  • Keynote, SSDBM Conference, Rennes, July 12, 2024
  • Keynote, SBBD Conference, Florianopolis, Brazil, October 14, 2024
• A. Joly
  • Invited keynote speaker at EGC 2024, Dijon, January 22-26, 2024
  • Invited speaker at a symposium at Collège de France entitled "Solutions to Monitor Plants, Pollinators and Their Interactions in a Changing World", Paris, May 23, 2024.
  • Invited panelist at a plenary session of the CESE (Conseil Economique, Social et Environnemental), Paris, September 23, 2024 (replay).

10.1.4 Leadership within the scientific community

• E. Pacitti: Member of the Steering Committee of the BDA conference.
• R. Akbarinia: Member of the Steering Committee of the BDA conference.
• : A. Joly
  • Scientific and Technical director of Pl@ntNet platform
  • Coordinator of the LifeCLEF international virtual lab

10.1.5 Scientific expertise

• C. Pradal: member of the INRAE evaluation committee CSS (Scientific Specialist Commission) in Plant Integrated Biology
• R. Akbarinia: member of the evaluation committee (section 27) of University of Montpellier.
• A. Joly:
  • GENCI expert committee (AI thematic)
  • AI2050 Early Career Fellowship
• P. Valduriez: consultant on big data for the Software Heritage project
• F. Masseglia: Inria International Chair selection committee

10.1.6 Research administration

• F. Masseglia: deputy scientific director of Inria for the domain "Perception, Cognition And Interaction", 50% of his time.
• R. Akbarinia: Scientific referent for research data at Inria branch of Montpellier; Member of Inria national commission for research data.
• E. Pacitti: manager of Polytech' Montpellier's International Relationships for the computer science department (100 students).
• P. Valduriez: scientific manager for the Latin America zone at Inria's Direction of Foreign Relationships (DRI) and scientific director of the Inria-Brasil strategic partnership.
• C. Pradal: Team leader with C. Granier of the PhenoMEn team of the AGAP Institute.
• A. Joly: co-manager of a Collaborative Doctoral Partnership between the EU Joint Research Centered of Ispra and the university of Montpellier

10.2.1 Teaching

Esther Pacitti:

• IG3: Database design, physical organization, 54h, level, L3, 50 students.
• IG4: Distributed Databases and NoSQL, 80h , level M1, 50 students.
• Large Scale Information Management (Iot, Recommendation Systems, Graph Databases), 27h, level M2, 20 students.
• Supervision of industrial projects
• Supervision of master internships.
• Supervision of computer science discovery projects.

10.2.2 Supervision

PhD & HDR:

• PhD in progress: Cesar Leblanc, Predicting biodiversity future trajectories through deep learning. Advisors: Alexis Joly, Maximilien Servajean, Pierre Bonnet.
• PhD in progress: Tanguy Lefort, Ambiguity of classification labels and expert feedback. Advisors: Joseph Salmon, Benjamin Charlier, Alexis Joly.
• PhD in progress: Kawtar Zaher, Novel class retrieval through interactive learning. Advisors: Olivier Buisson, Alexis Joly.
• PhD in progress: Matteo Contini, Multi-scale monitoring of coastal marine biodiversity. Advisors: Sylvain Bonhommeau, Alexis Joly.
• PhD in progress: Guillaume Coulaud, Anomaly Detection in Big Climate Data. Advisors: Reza Akbarinia, Audrey Brouillet, Florent Masseglia.
• PhD in progress: Loaï Gandeel, Automatic methods for spatio-temporal reconstruction of root architecture. Advisors: Reza Akbarinia, Romain Fernandez, Christophe Pradal.
• PhD in progress: Raphaël Benerradi, species trends estimation from citizen science data. Advisors: Christophe Botella, Alexis Joly, Maximilien Servajean
• PhD in progress: Théo Larcher, multi-scale species prediction. Advisors: Alexis Joly, Joseph Salmon, Pierre Bonnet, Marijn Van der Velde
• PhD in progress: Sébastien Gigot-Leandri, decision-oriented site occupancy models. Advisors: Alexis Joly, Maximilien Servajean, David Mouillot

10.2.3 Juries

Members of the team participated to the following PhD or HDR committees:

• R. Akbarinia (reviewer): Qi Fan, École Polytechnique.
• F. Masseglia: Anton Dolhopolov, Université de Montpellier.
• A. Joly: (i) as reviewer: Salim Khazem (Centrale supelec, 2024), Odile Peyron (University of Montpellier, 2024), (ii) as president of the jury: Letizia Lamperti (University of Montpellier, 2024), Ilyass Moummad (IMT Atlantique, 2024)

10.3.1 Specific official responsibilities in science outreach structures

• F. Masseglia: member of the strategic committee of Fondation Blaise Pascal
• A. Joly: member of the steering committee of Pl@ntNet citizen science platform

10.3.2 Productions (articles, videos, podcasts, serious games, ...)

• P. Valduriez
  • Interview Le Monde Informatique: "L'IA va-t-elle remplacer SQL"
  • Chiche (4 actions): Lycée La Mercy, Lycée Mermoz, Montpellier
• Hugo Gresse, Thomas Paillot: fête de la nature, Zoo de Lunaret, May 25, 2024
• A. Joly
  • Citizen science program La flore des cultures sous l'objectif de Pl@ntNet
  • Shooting of a 50-minute documentary for France 5 on gardening today (directed by Jean-Christophe Chatton, to be broadcasted in 2025)
  • Pl@ntNet's donation campaign 2024
• J. Salmon
  • fête de la science at university of Montpellier (carasciences)
  • bar des sciences, Faut il avoir peur de l'IA ?

International journals

• 15 articleO.Olivier Beaumont, L.Lionel Eyraud-Dubois, J.Julien Herrmann, A.Alexis Joly and A.Alena Shilova. Optimal checkpointing for heterogeneous chains: how to train deep neural networks with limited memory.ACM Transactions on Mathematical Software2024. In press. HALback to text
• 16 articleC.Christophe Botella, P.Pierre Gaüzère, L.Louise O'Connor, J.Julien Renaud, Y.Yue Dou, C.Catherine Graham, P.Peter Verburg, L.Luigi Maiorano and W.Wilfried Thuiller. Don't bite the hand that feeds you: Meta food webs help in the face of the Eltonian shortfall.Global Change Biology306June 2024, e17359HALDOI
• 17 articleS.Sherif Bouzaouia, M.Maxime Ryckewaert, D.Daphné Héran, A.Arnaud Ducanchez and R.Ryad Bendoula. Using Dynamic Laser Speckle Imaging for Plant Breeding: A Case Study of Water Stress in Sunflowers.Sensors24August 2024HALDOI
• 18 articleA. Y.Abel Yu Hao Chai, S. H.Sue Han Lee, F. S.Fei Siang Tay, P.Pierre Bonnet and A.Alexis Joly. Beyond supervision: Harnessing self-supervised learning in unseen plant disease recognition.Neurocomputing610December 2024, 128608HALDOIback to textback to text
• 19 articleJ.Jose Diaz-Olivares, R.Ryad Bendoula, W.Wouter Saeys, M.Maxime Ryckewaert, I.Ines Adriaens, X.Xinyue Fu, M.Matti Pastell, J.-M. J.Jean-Michel J. -M. Roger and B.Ben Aernouts. PROSAC as a selection tool for SO-PLS regression: A strategy for multi-block data fusion.Analytica Chimica Acta1319August 2024, 342965HALDOI
• 20 articleN.Neija Elvekjaer, L.Laura Martínez-Sanchez, P.Pierre Bonnet, A.Alexis Joly, M. L.Maria Luisa Paracchini and M.Marijn van Der Velde. Detecting flowers on imagery with computer vision to improve continental scale grassland biodiversity surveying.Ecological Solutions and Evidence5May 2024HALDOI
• 21 articleJ.Joaquim Estopinan, M.Maximilien Servajean, P.Pierre Bonnet, A.Alexis Joly and F.François Munoz. Mapping global orchid assemblages with deep learning provides novel conservation insights.Ecological Informatics81July 2024, 102627HALDOIback to text
• 22 articleC.César Leblanc, P.Pierre Bonnet, M.Maximilien Servajean, M.Milan Chytrý, S.Svetlana Aćić, O.Olivier Argagnon, A.Ariel Bergamini, I.Idoia Biurrun, G.Gianmaria Bonari, J. A.Juan Antonio Campos, R.Renata Ćušterevska, A.Andraž Čarni, M.Michele de Sanctis, J.Juergen Dengler, E.Emmanuel Garbolino, V.Valentin Golub, U.Ute Jandt, F.Florian Jansen, M.Maria Lebedeva, J. R.Jonathan Roger Michel Henri Lenoir, J. E.Jesper Erenskjold Moeslund, A.Aaron Pérez-Haase, R.Remigiusz Pielech, J.Jozef Šibík, Z.Zvjezdana Stančić, A.Angela Stanisci, G.Grzegorz Swacha, D.Domas Uogintas, K.Kiril Vassilev, T.Thomas Wohlgemuth and A.Alexis Joly. A deep-learning framework for enhancing habitat identification based on species composition.Applied Vegetation Science273August 2024, e12802HALDOIback to text
• 23 articleS. H.Sue Han Lee, Z. R.Zhe Rui Liaw, Y. H.Yu Hao Chai, S. L.Shien Lin Ng, P.Pierre Bonnet, H.Hervé Goëau and A.Alexis Joly. Revolutionizing Plant Pathogen Conservation: The Past, Present, and Future of AI in Preserving Natural Ecosystems.Biodiversity Information Science and Standards8July 2024, e133055HALDOI
• 24 articleT.Tanguy Lefort, B.Benjamin Charlier, A.Alexis Joly and J.Joseph Salmon. Identify ambiguous tasks combining crowdsourced labels by weighting Areas Under the Margin.Transactions on Machine Learning Research Journal2024HAL
• 25 articleT.Tanguy Lefort, B.Benjamin Charlier, A.Alexis Joly and J.Joseph Salmon. Peerannot: classification for crowdsourced image datasets with Python.Computo2024HALDOI
• 26 articleG.Gaétan Morand, A.Alexis Joly, T.Tristan Rouyer, T.Titouan Lorieul and J.Julien Barde. Predicting species distributions in the open ocean with convolutional neural networks.Peer Community Journal42024, e93HALDOIback to text
• 27 articleC.Corinne Robert, F.Francesco Accatino, A.Audrey Barbe, C.Carole Bedos, P.Pierre Benoit, C.Colette Bertrand, A.Amélie Bourceret, T.Thomas Da Costa, M.Magali Dahirel, C.Christian Fournier, T.Thibaut Griessinger, L.Lou Gauthier, L.Léa Grohens, F.Faustine Honore, M.Margaux Jacob, J.Jane Lecomte, L.Lucie Martin, E.Elliot Meunier, P.-A.Pierre-Antoine Précigout, C.Christophe Pradal, J.-E.Jean-Emmanuel Rougier and P.Pauline Smith. TRAVERSéES: Territorial levers and transition pathways for reducing pesticide use.Innovations Agronomiques962024, 25-37HALDOIback to text
• 28 articleR.Raphael Saldanha, R.Reza Akbarinia, M.Marcel Pedroso, V.Victor Ribeiro, C.Carlos Cardoso, E.Eduardo Peña, P.Patrick Valduriez and F.Fabio Porto. Zonal statistics datasets of climate indicators for Brazilian municipalities.Environmental Data Science3February 2024, e2HALDOI
• 29 articleR.Rebecca Salles, J.Janio Lima, R.Rafaelli Coutinho, E.Esther Pacitti, F.Florent Masseglia, R.Reza Akbarinia, C.Chao Chen, J. M.Jonathan M. Garibaldi, F. A.Fábio André Machado Porto and E. S.Eduardo S. Ogasawara. SoftED: Metrics for Soft Evaluation of Time Series Event Detection.Computers & Industrial Engineering2025. In press. HALDOIback to text

International peer-reviewed conferences

• 30 inproceedingsV.Vincent Berry, A.Arnaud Castelltort, B.Benoit Lange, J.Joan Teriihoania, C.Chouki Tibermacine and C.Catia Trubiani. Is it Worth Migrating a Monolith to Microservices? An Experience Report on Performance, Availability and Energy Usage.IEEE XploreICWS 2024 - IEEE International Conference on Web Services2024 IEEE International Conference on Web Services (ICWS)Shenzhen, ChinaIEEEJuly 2024, 944-954HALDOI
• 31 inproceedingsE.Eva Coindre, L.Laurine Chir, M.Maxime Ryckewaert, R.Romain Boulord, M.Mélyne Falcon, T.Thomas Laisné, G.Gaëlle Rolland, M.Maëlle Lis, L.Llorenç Cabrera-Bosquet, A.Agnes Doligez, T.Thierry Simonneau, B.Benoit Pallas, A.Aude Coupel-Ledru and V.Vincent Segura. Diversity of leaf functioning under water deficit in a large grapevine panel: high throughput phenotyping and genetic analyses.IVES Conference SeriesOpen GPB 2024 - Open Conference on Grapevine Physiology and BiotechnologyLogrono La Riora, SpainJuly 2024HALDOI
• 32 inproceedingsE.Eva Coindre, M.Maxime Ryckewaert, L.Laurine Chir, R.Romain Boulord, M.Mélyne Falcon, T.Thomas Laisné, G.Gaëlle Rolland, V.Virginie Bouckenooghe, M.Maëlle Lis, L.Llorenç Cabrera-Bosquet, A.Agnes Doligez, T.Thierry Simonneau, V.Virgilio Freitas, M.Miguel Thomas, B.Benoit Pallas, A.Aude Coupel-Ledru and V.Vincent Segura. NIRS as a high-throughput phenotyping tool for assessing the diversity of leaf functioning under water deficit in a large grapevine panel.Hélio SPIR25èmes rencontres HélioSPIRMontpellier, France2024, 23HAL
• 33 inproceedingsL. M.Lina M Estupinan Suarez, L.Laura Abraham, T.Tim Adriaens, L.Lissa Breugelmans, D. A.David A Clarke, P.Peter Desmet, S.Shawn Dove, K. T.Katelyn T Faulkner, M.Miguel Fernandez, L. A.Louise A Hendrickx, C.Cang Hui, A.Alexis Joly, S.Sabrina Kumschick, W.Ward Langeraert, M.Matilde Martini, J.Joe Miller, D.Damiano Oldoni, H.Henrique Pereira, C.Cristina Preda and Q.Quentin Groom. Biodiversity Data Cubes for Cross-Cutting Science and Policy.EGU 2024 abstractEGU 2024 - General Assembly of European Geosciences UnionVienna, AustriaNovember 2024HALDOI
• 34 inproceedingsR.Raphael de Freitas Saldanha, V.Victor Ribeiro, E.Eduardo Peña, M.Marcel Pedroso, R.Reza Akbarinia, P.Patrick Valduriez and F.Fabio Porto. Subset Models for Multivariate Time Series Forecast.IEEE XploreICDEW 2024 - IEEE 40th International Conference on Data Engineering WorkshopsInternational Conference on Data Engineering WorkshopsUtrecht, NetherlandsMay 2024, 86-90HALDOIback to text
• 35 inproceedingsH.Hervé Goëau, V.Vincent Espitalier, P.Pierre Bonnet and A.Alexis Joly. Overview of PlantCLEF 2024: multi-species plant identification in vegetation plot images.CLEF 2024 Working Notes - 25th Conference and Labs of the Evaluation Forum3740CEUR worshop proceedings187Grenoble, FranceSeptember 2024, 1978-1988HALback to text
• 36 inproceedingsA.Alexis Joly, L.Lukáš Picek, S.Stefan Kahl, H.Hervé Goëau, V.Vincent Espitalier, C.Christophe Botella, B.Benjamin Deneu, D.Diego Marcos, J.Joaquim Estopinan, C.César Leblanc, T.Théo Larcher, M.Milan Šulc, M.Marek Hrúz, M.Maximilien Servajean, J.Jiří Matas, H.Hervé Glotin, R.Robert Planqué, W.-P.Willem-Pier Vellinga, H.Holger Klinck, T.Tom Denton, A.Andrew Durso, I.Ivan Eggel, P.Pierre Bonnet and H.Henning Müller. LifeCLEF 2024 Teaser: Challenges on Species Distribution Prediction and Identification.Lecture notes in computer scienceECIR 2024 - 46th European Conference on Information RetrievalLNCS-14613Advances in Information Retrieval. ECIR 2024 - 46th European Conference on Information Retrieval, ECIR 2024, Glasgow, UK, March 24–28, 2024, Proceedings, Part VIGlasgow, United KingdomSpringer Nature SwitzerlandMarch 2024, 19-27HALDOIback to text
• 37 inproceedingsA.Alexis Joly, L.Lukáš Picek, S.Stefan Kahl, H.Hervé Goëau, V.Vincent Espitalier, C.Christophe Botella, D.Diego Marcos, J.Joaquim Estopinan, C.Cesar Leblanc, T.Théo Larcher, M.Milan Šulc, M.Marek Hrúz, M.Maximilien Servajean, H.Hervé Glotin, R.Robert Planqué, W.-P.Willem-Pier Vellinga, H.Holger Klinck, T.Tom Denton, I.Ivan Eggel, P.Pierre Bonnet and H.Henning Müller. Overview of LifeCLEF 2024: Challenges on Species Distribution Prediction and Identification.Lecture notes in computer scienceCLEF 2024 - 15th International Conference of the Cross-Language Evaluation Forum for European LanguagesLNCS-14959Experimental IR Meets Multilinguality, Multimodality, and Interaction : 15th International Conference of the CLEF Association, CLEF 2024, Grenoble, France, September 9–12, 2024, Proceedings, Part IIGrenoble, FranceSpringer Nature SwitzerlandSeptember 2024, 183-207HALDOIback to text
• 38 inproceedingsS.Stefan Kahl, T.Tom Denton, H.Holger Klinck, V.Vijay Ramesh, V.Viral Joshi, M.Meghana Srivathsa, A.Akshay Anand, C.Chiti Arvind, H.Harikrishnan Cp, S.Suyash Sawant, H.Hervé Glotin, H.Hervé Goëau, W.-P.Willem-Pier Vellinga, R.Robert Planqué and A.Alexis Joly. Overview of BirdCLEF 2024: Acoustic Identification of Under-studied Bird Species in the Western Ghats.CLEF 2024 Working Notes - 25th Conference and Labs of the Evaluation Forum3740CEUR workshop proceedingsGrenoble, FranceSeptember 2024, 1948-1957HALback to text
• 39 inproceedingsB.Benoit Lange, R.Reza Akbarinia and F.Florent Masseglia. A One-Health Platform for Antimicrobial Resistance Data Analytics.CIKM 2024 - 33rd ACM International Conference on Information and Knowledge ManagementCIKM '24: Proceedings of the 33rd ACM International Conference on Information and Knowledge ManagementBoise, United StatesOctober 2024, 5230-5233HALDOIback to text
• 40 inproceedingsJ. Z.Jerad Zherui Liaw, A. Y.Abel Yu Hao Chai, S. H.Sue Han Lee, P.Pierre Bonnet and A.Alexis Joly. Can Language Improve Visual Features For Distinguishing Unseen Plant Diseases?Lecture Notes in Computer ScienceICPR 2024 - 27th International Conference on Pattern Recognition15330Lecture Notes in Computer ScienceKolkata, IndiaSpringer Nature SwitzerlandDecember 2024, 296-311HALDOIback to textback to text
• 41 inproceedingsJ.Janio Lima, L.Lucas Tavares, E.Esther Pacitti, J. E.João Eduardo Ferreira, I.Ismael Santos, I.Isabela Guimaraes Siqueira, D.Diego Carvalho, F.Fabio Porto, R.Rafaelli Coutinho and E.Eduardo Ogasawara. Online Event Detection in Streaming Time Series: Novel Metrics and Practical Insights.IJCNN 2024 - International Joint Conference on Neural NetworksYokoama, JapanJune 2024, 1-8HAL
• 42 inproceedingsJ.Ji Liu, T.Tianshi Che, Y.Yang Zhou, R.Ruoming Jin, H.Huaiyu Dai and P.Patrick Valduriez. AEDFL: Efficient Asynchronous Decentralized Federated Learning with Heterogeneous Devices.SDM 2024 - SIAM International Conference on Data MiningHouston, TX, United StatesSociety for Industrial and Applied MathematicsJanuary 2024, 833-841HALDOIback to text
• 43 inproceedingsJ.Ji Liu, J.Jiaxiang Ren, R.Ruoming Jin, Z.Zijie Zhang, Y.Yang Zhou, P.Patrick Valduriez and D.Dejing Dou. Fisher Information-based Efficient Curriculum Federated Learning with Large Language Models.EMNLP 2024 - Conference on Empirical Methods in Natural Language ProcessingMiami, Fl, United StatesOctober 2024, 1-27HALback to text
• 44 inproceedingsE.Ellen Paixão Silva, H.Helga Balbi, E.Esther Pacitti, F.Fabio Porto, J. A.Joel A. F. dos Santos and E. S.Eduardo S. Ogasawara. Cutoff Frequency Adjustment for FFT-Based Anomaly Detectors.SBBD 2024 - Simpósio Brasileiro de Banco de DadosFlorianapolis, BrazilOctober 2024, 1-5HALback to text
• 45 inproceedingsL.Lukáš Picek, C.Christophe Botella, M.Maximilien Servajean, C.César Leblanc, R.Rémi Palard, T.Théo Larcher, B.Benjamin Deneu, D.Diego Marcos, P.Pierre Bonnet and A.Alexis Joly. GeoPlant: Spatial Plant Species Prediction Dataset.NeurIPS 2024 - 38th Conference on Neural Information Processing SystemsTrack on Datasets and BenchmarksVancouver, CanadaDecember 2024HALback to text
• 46 inproceedingsL.Lukáš Picek, C.Christophe Botella, M.Maximilien Servajean, C.César Leblanc, R.Remi Palard, T.Théo Larcher, B.Benjamin Deneu, D.Diego Marcos, J.Joaquim Estopinan, P.Pierre Bonnet and A.Alexis Joly. Overview of GeoLifeCLEF 2024: Species Composition Prediction with High Spatial Resolution at Continental Scale using Remote Sensing.CLEF 2024 Working Notes - 25th Conference and Labs of the Evaluation Forum3740CEUR workshop proceedings186Grenoble, FranceCEURSeptember 2024, 1966-1977HALback to text
• 47 inproceedingsL.Lukáš Picek, M.Milan Šulc and J.Jiří Matas. Overview of FungiCLEF 2024: Revisiting Fungi Species Recognition Beyond 0-1 Cost.CLEF 2024 Working Notes - 25th Conference and Labs of the Evaluation Forum3740CEUR workshop proceedings185Grenoble, FranceSeptember 2024, 1958-1965HALback to text
• 48 inproceedingsM.Michel Reis, R.Rebecca Salles, G.Geraldo Xexeo, R.Rafaelli Coutinho and E.Eduardo Ogasawara. Matching Detections to Events in Time Series.SBBD 2024 - Simpósio Brasileiro de Banco de DadosFlorianapolis, BrazilOctober 2024, 1-6HALback to text
• 49 inproceedingsR. A.Rodrigo A. P. Silva, W.Wesley Ferreira, E.Esther Pacitti, Y.Yuri Frota and D.Daniel de Oliveira. A Heuristic for Executing Confidentiality-Constrained Workflows in Containerized Environments.SBBD 2024 - Simpósio Brasileiro de Banco de DadosFlorianapolis, BrazilOctober 2024, 1-13HALback to text
• 50 inproceedingsR. A.Rodrigo A. P. Silva, Y. Y.Yuri Y. Frota, D.Daniel de Oliveira and E.Esther Pacitti. Assegurando a Confidencialidade de Dados de Workflows Executados em Nuvens de Computadores: Abordagens Heurísticas e Exatas.SBPO 2024 - Simpósio Brasileiro de Pesquisa OperacionalFortaleza, BrazilNovember 2024, 1-12HALback to text

National peer-reviewed Conferences

• 51 inproceedingsA.Axel Dubar, T.Tanguy Lefort and J.Joseph Salmon. peerannot: A framework for label aggregation in crowdsourced datasets.JDS 2024 - 55es Journées de StatistiqueRecueil de l'ensemble des résumés longs est téléchargeableBordeaux, FranceMay 2024HAL
• 52 inproceedingsT.Tanguy Lefort, A.Antoine Affouard, P.Pierre Bonnet, B.Benjamin Charlier, A.Alexis Joly and J.Joseph Salmon. Cooperative learning of Pl@ntNet’s Artificial Intelligence algorithm using label aggregation.Recueil des résumés longs : 55èmes Journées de Statistique de la SFdSJDS 2024 - 55es Journées de StatistiqueRecueil des résumés longs : 55èmes Journées de Statistique de la SFdSBordeaux, FranceMay 2024, 35-44HAL

Conferences without proceedings

• 53 inproceedingsT.Tanguy Lefort, B.Benjamin Charlier, A.Alexis Joly and J.Joseph Salmon. Weighted majority vote using Shapley values in crowdsourcing.CAp 2024 - Conférence sur l'Apprentissage AutomatiqueLille, FranceMay 2024HAL

Scientific books

• 54 bookE. S.Eduardo S. Ogasawara, R.Rebecca Salles, F.Fabio Porto and E.Esther Pacitti. Event Detection in Time Series.Synthesis Lectures on Data Management (SLDM)SpringerFebruary 2025, 1-178HALback to text

Scientific book chapters

• 55 inbookR.Rocío Zorrilla, E.Eduardo Ogasawara, P.Patrick Valduriez and F.Fábio Porto. A Data-Driven Model Selection Approach to Spatio-Temporal Prediction.LNCS-14790Transactions on Large-Scale Data- and Knowledge-Centered Systems LVI : Special Issue on Data Management - Principles, Technologies, and ApplicationsLecture Notes in Computer Science. Transactions on Large-Scale Data- and Knowledge-Centered SystemsJuly 2024, 98-118HALDOIback to text

Edition (books, proceedings, special issue of a journal)

• 56 proceedingsR.Reza Akbarinia, T.Tristan Allard and A.Angela Bonifati, eds. Actes de la conférence BDA 2023, Montpellier.BDA 2023 - 39ème Conférence sur la Gestion de Données Principes Technologies et ApplicationsJune 2024HAL
• 57 periodicalA.Abdelkader Hameurlain, A. M.A Min Tjoa, R.Reza Akbarinia and A.Angela Bonifati, eds. Transactions on Large-Scale Data- and Knowledge-Centered Systems; Vol. LVI, LNCS 14790. Special Issue on Data Management - Principles, Technologies, and Applications.14790Lecture Notes in Computer Science2024HALDOI

Reports & preprints

• 58 misc T.Tanguy Lefort, A.Antoine Affouard, B.Benjamin Charlier, J.-C.Jean-Christophe Lombardo, M.Mathias Chouet, H.Hervé Goëau, J.Joseph Salmon, P.Pierre Bonnet and A.Alexis Joly. Cooperative learning of Pl@ntNet's Artificial Intelligence algorithm: how does it work and how can we improve it? June 2024 HAL DOI back to text

Other scientific publications

• 59 articleE.Emmanuelle Porcher, P.Pierre Bonnet, C.Christian Damgaard, P.Pieter de Frenne, N.Nicolas Deguines, B.Bodil Ehlers, J.Jérôme Frei, M.María García, C.Clément Gros, U.Ute Jandt, A.Alexis Joly, G.Gabrielle Martin, D.Denis Michez, O.Oliver Pescott, T.Tobias Roth and D.Donald Waller. Can we harmonize the monitoring of plants and pollinators?New Phytologist2441September 2024, 39-42HALDOI