Robust Learning and Reasoning for Complex Event Forecasting

Informations projet

EVENFLOW

N° de convention de subvention: 101070430

DOI

10.3030/101070430

Projet clôturé

Date de signature de la CE 20 Juin 2022

Date de début 1 Octobre 2022

Date de fin 31 Decembre 2025

Financé au titre de

• Digital, Industry and Space

Coût total

€ 3 431 000,00

Contribution de l’UE

€ 3 431 000,00

Coordonné par

NETCOMPANY SA
Belgium

CORDIS fournit des liens vers les livrables publics et les publications des projets HORIZON.

Les liens vers les livrables et les publications des projets du 7e PC, ainsi que les liens vers certains types de résultats spécifiques tels que les jeux de données et les logiciels, sont récupérés dynamiquement sur OpenAIRE .

Livrables

Documents, reports (7)

Ethics Manual on Trustworthy Neuro-symbolic Learning for Complex Event Forecasting

A report that will a) explain how all data intended to be used for both training and/or validating the project’s techniques and models are relevant, high quality, and limited to the purposes at hand, at both micro and macro level, b) assess whether the neuro-symbolic models and their use in the context of the use cases have any detrimental impact on humans, namely in terms of safety (applicable for all use cases) and contingently discrimination (applicable in use case 2), c) provide adequate information regarding the system's quality, following a risk-based approach and classifying the systems’ risk levels.

Architecture Design and Integrated System Specification (PU version)

A report on the architecture design, software specifications for the EVENFLOW connectors and best practices towards the development of the integrated prototype.

Data handling, Requirements Analysis & Scenario Definition

This deliverable will present the data handling process, the requirements analysis and the scenario definitions for the use cases.

Interim Version of Verification and Scalability Techniques

First version of robustness and efficiency algorithms and tools that support EVENFLOW’s online learning and reasoning approach.

Interim Version of Online Neuro-Symbolic Learning & Reasoning Techniques

First version of online neuro-symbolic learning and reasoning, explainability and data programming techniques.

Plan for Dissemination and Exploitation Including Communication Activities

This deliverable will detail the plan for the project dissemination and exploitation strategy to be adopted throughout the project lifetime. Additionally, it will include a report on related projects and actions that EVENFLOW will cooperate with, in order to approach stakeholders.

Interim Use Case Evaluation

This deliverable will present the results and feedback from the interim evaluation of the tools and the EVENFLOW platform in the three use cases.

Data Management Plan (1)

Data Management Plan

A report that will a) provide an overview of the datasets generated and used in the context of the project and the use cases, b) outline how data will be generated and/or collected and processed, subject to all relevant regulations, namely the GDPR, the Data Governance Act and the Proposal for the AI Regulation, c) assess, on top of the methodology and standards to be introduced, whether and how this data will be shared as FAIR and how it will be curated and preserved, in direct alignment with the ethics requirements.

Websites, patent fillings, videos etc. (1)

Project Presentation and Website

This deliverable contains a fact sheet providing an overview of the project, following the EC standards and template, the project presentation and the project website, which will be the main dissemination and exploitation channel for promoting and communicating the project and relevant activities and achievements.

Publications

Conference proceedings (16)

A Novel Reverse Random Hyperplane Projection Scheme and Its Effect on Mining Sensor Streams

Auteurs: Antonios Skevis, George Klioumis, Nikos Giatrakos
Publié dans: 2024 IEEE International Conference on Big Data (BigData), 2025
Éditeur: IEEE
DOI: 10.1109/BIGDATA62323.2024.10825198

NeuroFlinkCEP: Neurosymbolic Complex Event Recognition Optimized across IoT Platforms

Auteurs: Ourania Ntouni, Dimitrios Banelas, Nikos Giatrakos
Publié dans: Proceedings of the VLDB Endowment, Numéro 18, 2025, ISSN 2150-8097
Éditeur: Association for Computing Machinery (ACM)
DOI: 10.14778/3750601.3750670

Neuro-symbolic Complex Event Recognition in Autonomous Driving

Auteurs: Tatiana Boura, Nikos Katzouris
Publié dans: 2025
Éditeur: Creative Commons license

Optimizing Resource Allocation for Tumor Simulations over HPC Infrastructures

Auteurs: Streviniotis, Errikos; Giatrakos, Nikos; Kotidis, Yannis; Ntiniakou, Thaleia; Ponce de Leon, Miguel;
Publié dans: 2023
Éditeur: IEEE
DOI: 10.1109/dsaa60987.2023.10302484

An Empirical Evaluation of Explainable AI Approaches

Auteurs: Ioannis T. Christou, John Soldatos, Pantelis Lappas
Publié dans: 2025 21st International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT), 2025
Éditeur: IEEE
DOI: 10.1109/DCOSS-IOT65416.2025.00159

Feature Selection via Minimal Covering Sets for Industrial Internet of Things Applications

Auteurs: Christou, Ioannis T.; Soldatos, John; Papadakis, Thanassis; Gutierrez-Rojas, Daniel; Nardelli, Pedro;
Publié dans: 2023
Éditeur: IEEE
DOI: 10.1109/dcoss-iot58021.2023.00092

SSTRESED: Scalable Semantic Trajectory Extraction for Simple Event Detection over Streaming Movement Data

Auteurs: Giatrakos, Nikos
Publié dans: International Symposium on Temporal Representation and Reasoning (TIME), Numéro 30, 2023
Éditeur: International Symposium on Temporal Representation and Reasoning (TIME)
DOI: 10.4230/LIPIcs.TIME.2023.15

NeSyA: Neurosymbolic Automata

Auteurs: Manginas, Nikolaos; Paliouras, George; De Raedt, Luc
Publié dans: Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence, 2025
Éditeur: Springer Nature
DOI: 10.24963/IJCAI.2025/662

A Scalable Approach to Probabilistic Neuro-Symbolic Robustness Verification

Auteurs: Vasileios Manginas ~Vasileios_Manginas1 , Nikolaos Manginas, Edward Stevinson, Sherwin Varghese, Nikos Katzouris, Georgios Paliouras, Alessio Lomuscio
Éditeur: Open Review

SuBiTO: Synopsis-based Training Optimization for Continuous Real-Time Neural Learning over Big Streaming Data

Auteurs: Errikos Streviniotis, George Klioumis, Nikos Giatrakos
Publié dans: Proceedings of the AAAI Conference on Artificial Intelligence, Numéro 39, 2025, ISSN 2374-3468
Éditeur: Association for the Advancement of Artificial Intelligence (AAAI)
DOI: 10.1609/AAAI.V39I28.35371

Answer Set Automata: A Learnable Pattern Specification Framework for Complex Event Recognition

Auteurs: Katzouris, Nikos; Paliouras, Georgios;
Publié dans: Leibniz International Proceedings in Informatics (LIPIcs), Numéro 278, 2023
Éditeur: Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2023)
DOI: 10.4230/LIPIcs.TIME.2023.17

Proactive Streaming Analytics at Scale: A Journey from the State-of-the-art to a Production Platform

Auteurs: Giatrakos, Nikos; Alevizos, Elias; Deligiannakis, Antonios; Kinkenberg, Ralf; Artikis, Alexander
Publié dans: 2023
Éditeur: CIKM ’23
DOI: 10.1145/3583780.3615293

Expressive Losses for Verified Robustness via Convex Combinations

Auteurs: Alessandro De Palma, Rudy Bunel, Krishnamurthy Dvijotham, M. Pawan Kumar, Robert Stanforth, Alessio Lomuscio
Publié dans: 2024
Éditeur: ICLR via OpenReview

How to Make your Duck Fly: Advanced Floating Point Compression to the Rescue

Auteurs: Liakos, Panagiotis; Papakonstantinopoulou, Katia; Bruineman, Thijs; Raasveldt, Mark; Kotidis, Yannis
Publié dans: 2024
Éditeur: EDBT
DOI: 10.5281/zenodo.10815563

Data-driven Synchronization Protocols for Data-parallel Neural Learning over Streaming Data

Auteurs: George Klioumis, Nikos Giatrakos
Publié dans: 2024 IEEE International Conference on Big Data (BigData), 2025
Éditeur: IEEE
DOI: 10.1109/BIGDATA62323.2024.10825830

Model Predictive Control Based Reference Generation for Optimal Proportional Integral Derivative Control

Auteurs: Fatos Gashi, Khalil Abuibaid, Martin Ruskowski, Achim Wagner
Publié dans: 2024 32nd Mediterranean Conference on Control and Automation (MED), 2024
Éditeur: IEEE
DOI: 10.1109/MED61351.2024.10566273

Peer reviewed articles (3)

Nature Communications

Auteurs: Núñez-Carpintero, I; O’Connor, E; Rigau, M; Bosio, M; Spendiff, S; Azuma, Y; Topf, A; Thompson, R; A.C.’t Hoen, P; Chamova, T; Tournev, I; Guergueltcheva, V; Laurie, S; Beltran, S; Capella, S; Cirillo, D; Lochmüller, H; Valencia, A;
Publié dans: Nature Communications, Numéro 15, 2023, ISSN 2041-1723
Éditeur: Nature Publishing Group
DOI: 10.1101/2023.01.19.524736,10.1038/s41467-024-45099-0

RATS: A resource allocator for optimizing the execution of tumor simulations over HPC infrastructures

Auteurs: Errikos Streviniotis, Nikos Giatrakos, Yannis Kotidis, Thaleia Ntiniakou, Miguel Ponce de Leon
Publié dans: Information Systems, Numéro 132, 2025, ISSN 0306-4379
Éditeur: Elsevier BV
DOI: 10.1016/J.IS.2025.102538

And synopses for all: A synopses data engine for extreme scale analytics-as-a-service. Information Systems

Auteurs: Antonios Kontaxakis; Nikos Giatrakos; Dimitris Sacharidis; Antonios Deligiannakis
Publié dans: Information Systems, Numéro Volume 116, 2023, ISSN 0306-4379
Éditeur: Science Direct
DOI: 10.1016/j.is.2023.102221

Other (2)

From Complexity to Clarity: Evaluating Explainability of Biomedical Machine Learning Models

Auteurs: González Mallo, Marta; Valencia, Alfonso; Cirillo, Davide
Publié dans: 2024
Éditeur: ISMB/ECCB 2023
DOI: 10.5281/zenodo.10564978

Adaptive and Scalable Multi-Mobile-Robot Simulation

Auteurs: Dudhagara, S., Blumhofer, B., Ruskowski, M., & Wagner, A.
Publié dans: 2025
Éditeur: ZENODO
DOI: 10.5281/ZENODO.18469379