Openfund

CORDIS · HumAIne

Hybrid Human-AI Decision Support for Enhanced Human Empowerment in Dynamic Situations

Код
HumAIne
Програма
HORIZON
Процедура
HORIZON-CL4-2022-HUMAN-02-01
Статус
SIGNED
Период
01.10.2023 – 30.09.2026
Обща стойност (целият консорциум)7 658 375 €
Принос на ЕС (целият консорциум)7 658 375 €
Принос на ЕС за българските участници208 750 €

Официални данни · CORDIS · синхронизирано 17.10.2026 10:16 · официален запис

Участници

ОрганизацииРоляДържаваБезвъзмездна помощ
GFT ITALIA SRLкоординаторIT725 000 €
INSTITUT JOZEF STEFANучастникSI838 375 €
UNIVERSITY OF PIRAEUS RESEARCH CENTERучастникEL594 375 €
DEUTSCHES FORSCHUNGSZENTRUM FUR KUNSTLICHE INTELLIGENZ GMBHучастникDE587 500 €
UBITECH ENERGYучастникBE549 656 €
IBM ISRAEL - SCIENCE AND TECHNOLOGY LTDучастникIL536 250 €
УНИВЕРСИТЕТ ГРЬОНИНГЕНучастникNL535 000 €
INNOVATION SPRINTучастникBE530 000 €
INNOV-ACTS LIMITEDучастникCY456 250 €
UNPARALLEL INNOVATION LDAучастникPT430 000 €
LEANXCALE SLучастникES366 594 €
EBIT S.R.L.участникIT362 500 €
CHAROKOPEIO PANEPISTIMIOучастникEL334 375 €
IMMERSIVELIVES LDAучастникPT325 000 €
INNEUROPE INITIATIVE S.L.участникES278 750 €
VILABS BG LTDучастникBG208 750 €
QLECTOR, RAZVOJ CELOVITIH RESITEV ZA PAMETNE TOVARNE DOOтрета странаSI0 €
NOVOVILLE LIMITEDасоцииран партньорUK

Описание

HumAIne will research, develop, validate and promote a novel operating system for Human-AI collaboration, which will enable the development of advanced decision making applications in dynamic, unstructured environments in different industrial sectors. The HumAIne OS will empower AI solution integrators to implement Human-AI collaboration systems that outperform AI systems and humans when working in isolation. HumAIne’s developments will be integrated into a single OS platform, which will coordinate four interwind components offering Active Learning (AL), Neuro-Symbolic Learning, Swarm Learning (SL) as well eXplainable AI (XAI) capabilities. These advanced AI paradigms are ideal for exploiting true Human-AI collaboration since, in each of them, the worker is the key actor with complete control and understanding of the performed operations. AL enables the development of effective Human-in-the-Loop systems that involve humans when AI faces increased uncertainty. Neuro-Symbolic Learning combines DL with semantics and rules to complete highly complex tasks with high accuracy while requiring considerably less training data than current AI models. Advanced XAI models will be made available, providing explanations of models’ predictions while considering the global context instead of just analysing the feature importance of a single AI model. HumAIne’s XAI will provide guidance to humans to enable the timely optimisation of AL and SL models where human participants provide feedback dynamically as well as fine-tuning of Neuro-Symbolic models. The platform will handle various types of structured and unstructured data, including inputs from humans that will be semantically correlated through ontologies, knowledge graphs, and semantic interoperability. HumAIne will complement its platform with complementary resources (e.g., training) and will be build a vibrant community of interested parties around it, to drive exploitation and wider use of the project's results.

Официални данни · CORDIS · синхронизирано 17.10.2026 10:16 · официален запис · Лиценз: © European Union, CC BY 4.0 (Commission reuse policy).