Events

The second E-pi periodic review meeting with the project officer, experts, and all E-pi partners was successfully completed at Oxford Brookes University on November 6, 2024.

The seventh E-pi consortium meeting was hosted by Oxford Brookes University on November 5, 2024, in Oxford. All partners were engaged in updating the progress on the work packages and preparing for the second E-pi periodic review meeting.

The ROAD++ @ECCV 2024 workshop, co-organised by OBU and its partners, was held on September 29 2024. The goal of ROAD++ workshop is twofold: firstly, to bring to the forefront of research in autonomous driving on the topic of situational awareness. Secondly, to present ROAD++: a combination of four datasets from different domains including ROAD (UK), ROAD-Waymo (USA), ROAD-UAE (UAE), and TACO (CARLA simulation). For more information, kindly visit the website.

The 2024 edition of the biennial SIPTA School on Imprecise Probabilities took place from Monday 12 to Friday 16 August 2024, and provided the participants with the opportunity to get introduced to the basics of various topics in this theory, both theoretical and applied to different disciplines, by leading experts in the field. Mr Adam Faze (PhD student) participate in the school and Dr. Keivan Shariatmadar from KUL as the member of Organising Committee and Advisory Member at SIPTA summer school. For more information, kindly visit the website.

The sixth E-pi consortium meeting was hosted by KU Leuven on May 28 2024 in Leuven, Belgium. All partners were engaged in the progress update on the work packages and the technical brainstorming sessions.

The ROAD-R competition was co-organised by OBU and its partners in Oxford, Imperial College as part of NeurIPS 2023 and was aimed to address two key limitations of deep learning models: data dependency and compliance with logical requirements in autonomous driving. For more information, kindly visit the website.

The ROAD++ challenge and workshop was organised by OBU (and others) as part of ICCV 2023. There, we introduced the ROAD-Waymo dataset, which tackles the diversity and scale issues of ROAD, building on the Waymo-Open dataset. For more information, kindly visit the workshop's website.

The E-pi consortium organized a workshop entitled with E-pi: Re-thinking Uncertainty and AI, which took place on 22 September 2023, TU Delft, the Netherlands. This workshop aims to discuss the current state, limitations, and future perspectives on the foundations of uncertainty and AI. It gathered voices from artificial intelligence, decision making, and engineering to discuss handling uncertainty in a principled, but practical manner. For more information, kindly visit the workshop's website.

The fifth E-pi consortium meeting, held in person, was hosted by TU Delft on September 21 2023 in Delft, the Netherlands. All partners participated actively in the progress update on the Work Package(s) and the technical brainstorming sessions.

The E-pi UAI 2023 workshop, focusing on epistemic uncertainty in artificial intelligence, took place on August 4, Pittsburgh, USA. The workshop was co-organised by Fabio Cuzzolin (OBU), Matthijs Spaan (TUD), Keivan Shariatmadar (KUL), Maryam Sultana (OBU), Kaizheng Wang (KUL), and Shirren KUdukkil Manchingal (OBU). For this workshop, we accepted 16 non-archival papers covering a wide range of topics. Four speakers, Prof. Gert de Cooman from Ghent University, Prof. Yarin Gal from Oxford University, Dr. Marco Zaffalon from IDSIA, and Prof. Aaditya Ramdas from Carnegie Mellon University were also invited to provide insightful technical talks and foster meaningful discussions on uncertainty in artificial intelligence. For more information, kindly visit the workshop's website E-pi UAI.

The forth E-pi consortium meeting, held in person, was hosted by Oxford Brookes University on December 6 2022 in Headington, Oxford. During the meeting, PhD researcher Shireen Kudukkil Manchingal presented her recent research on Random-set CNN for image classification. All partners participated actively in the progress update on the Work Package(s) and the technical brainstorming sessions.

The third E-pi consortium meeting, held personally, was hosted by KU Leuven and organised by Dr Keivan Shariatmadar on May 18 2022 in KU Leuven, Campus Bruges, Belgium. A technical presentation on ''Safe Reinforcement Learning'' given by Dr. Matthijs Spaan (TU Delft), the project update report, the technical discussion, and a laboratory tour were among the activities.

The E-pi periodic technical and financial report with project officer and experts was conducted on May 11 2022, virtually.

The second E-pi consortium meeting, held virtually, was organised by Prof Neil Yorke-Smith and Pascal van der Vaart (TU Delft) on 8 December 2021. All partners had the opportunity to present their Work Package(s), including the associated Deliverables and Milestones. Meanwhile, a technical talk on "Imprecise Decision Theory" was given by Dr Keivan Shariatmadar (KU Leuven).

The goal of this workshop is to put to the forefront of the research in autonomous driving the topic of situation awareness, intended as the ability to create semantically useful representations of dynamic road scenes in terms of the notion of ‘road event’, itself inspired by the central computer vision notion of ‘action’. The workshop also aims to stimulate a change of paradigm in the field of action detection, by shifting the focus from the objects/actors themselves and their appearance to what they do and the meaning of their behaviour, as the concept of action is here extended to apply to human-operated machinery as an extension of the human mind.

The new Epistemic AI EU Future Emerging Technologies (FET) project has started on March 1 2021.

The kickoff meeting took place on Monday 15 March in a virtual fashion, and hosted as speakers Prof Fabio Cuzzolin and Dr Andrew Bradley (Oxford Brookes), Dr Keivan Shariatmadat (KU Leuven), Prof Neil Yorke-Smith (TU Delft), Prof Hans Hallez (KU Leuven), Prof David Moens (KU Leuven),  Dr Matthijs Spaan and Dr Julian Kooij (TU Delft).

The project is coordinated by Prof Fabio Cuzzolin and has a duration of 4 years.