Objective
Fast, accurate forecasting of spatiotemporal data is needed in critical industrial domains such as energy (prediction of spatiotemporal patterns in renewable generation, usage and traffic) as well as in public policy. The task is so challenging in scale and scope however as to have been confined mainly to research, while past prize competitions have been limited to forecasts of single dimensional values. Building on our proven success in numerous prize-driven past data challenges, attracting hundreds of participants, we aim to compile and test data grounded on large-scale open European datasets and including specially prepared grid traffic data from Europe’s largest Transportation System Operator. The competition evaluates forecasting algorithms on a cloud platform, tracking accuracy and computational efficiency. Emphasizing cross-specialization knowledge transfer and openness to novel technologies which may spring from different subsectors, we aim to build a platform allowing for coopetitions: the ad-hoc coalescence of competing teams during a challenge aimed at forming sustainable partnerships past the prize scheme itself. We will provide comprehensive documentation for a freely extensible open-source cloud-based specialized computing platform (assembling existing, well tested tools) allowing automated evaluation and feedback as in our latest competitions, but scaled to big data needs. We aim to test this platform and provide baseline results in a smaller scale mini-competition (hackathon). Thus we shall lay the groundwork for a larger prize competition in which evaluation data for predictions may arrive in real or near-real time. We also aim to use our wide contacts with industry, domain and data experts and past participants and winners in order to organize focused meetings of panels to refine value chains in data and algorithms as well as conference workshops, talks and newsletters dedicated to widely advertising challenges to past and new participants.
Fields of science
- social sciencespolitical sciencespolitical policiespublic policies
- engineering and technologyenvironmental engineeringenergy and fuelsrenewable energy
- natural sciencescomputer and information sciencesdata sciencebig data
- natural sciencesearth and related environmental sciencesatmospheric sciencesclimatology
- natural sciencescomputer and information sciencesartificial intelligencemachine learningdeep learning
Programme(s)
Topic(s)
Funding Scheme
CSA - Coordination and support action
Coordinator
80686 Munchen
Germany
See on map
Participants (4)
13284 Marseille
See on map
08007 Barcelona
See on map
Legal entity other than a subcontractor which is affiliated or legally linked to a participant. The entity carries out work under the conditions laid down in the Grant Agreement, supplies goods or provides services for the action, but did not sign the Grant Agreement. A third party abides by the rules applicable to its related participant under the Grant Agreement with regard to eligibility of costs and control of expenditure.
08193 Bellaterra Cerdanyola Del Valles
See on map
92073 Paris La Defense Cedex
See on map