16/04/2021
Marie Skłodowska-Curie Actions

MSCA- Postdoctoral Fellowship: Deep learning models for temporal data for predictive models


  • OFFER DEADLINE
    30/06/2021 12:00 - Europe/Athens
  • EU RESEARCH FRAMEWORK PROGRAMME
    HE / MSCA
  • LOCATION
    Spain, SEVILLA
  • ORGANISATION/COMPANY
    Universidad Pablo de Olavide
  • DEPARTMENT
    OTRI
  • LABORATORY
    Department of Computer Science

The Pablo de Olavide University in Seville offers positions for researchers willing to implement a 2-3 years project based in Spain within the framework of the Horizon Europe Marie Skłodowska Curie postdoctoral fellowships programme (MSCA-PF-2021). All top candidates having the approval of a supervisor at UPO will have the possibility of receiving specialized training in writing a successful proposal for the next MSCA-PF call. You can apply if at the time of the call deadline for the submission of MSCA-PF proposals (September, 2021): 1) you are in the possession of a PhD and you have not obtained your PhD more than 8 years ago; 2) you comply with the mobility rule of the MSCA; 3) you choose the Pablo de Olavide University as your Host Institution. We will be happy to support highly qualified candidates who are interested in any area of our research and are eligible for the MSCA funding. Please contact us by email to get all the details about available scientific supervisors and the procedure to benefit from the training on proposal writing.

Brief description of the Centre / Research Group:

Our research group is located at Pablo de Olavide University in the city of Seville in the south of Spain and is mainly focused on big data, time series, and machine learning. We participate in a great number of multi-disciplinary research projects, and closely collaborate with both private firms and international groups from other universities, achieving in all of them a high level of publications in indexed journals and conferences. Currently, they are working on the big data streaming and deep learning. In big data streaming, the group is designing algorithms for forecasting tasks in real time. In deep learning, the group is working in the fusion of deep learning models for vision artificial and forecasting tasks.

In terms of human resources, the group has 6 PhDs on staff at the Pablo de Olavide University. In addition, it has 7 researchers under contract (3 of them already has a PhD), all of them experts in artificial intelligence and big data. All of this means that the team can tackle large volumes of work with very competitive response times.

In terms of its own infrastructures, the group has a cluster of 9 physical machines (16 GB RAM, 1.2 TB HD-SSD, 3.2 GHz, 12 cores each). It also has a contract with a private company for the use of cloud infrastructures, so it provides its own equipment and makes it available to projects, if necessary.

Project description:

In the field of time data, open global competitions, known as M competitions, are organized with the intention of evaluating and comparing the accuracy of different prediction methods proposed by researchers and companies. In the last competition held in 2018, the M4 competition, which consisted of the prediction of one hundred thousand time series with different measurement frequencies and different time horizons, the method that obtained the best predictions was a hybrid prediction method between a neural network and a statistical model proposed by Uber, and the second best method was a combination of a machine learning model and a statistical model proposed by a researcher at the University of La Coruña. These results obtained on such a large amount of temporal data, together with the ability to learn hidden relationships in the data that deep learning has demonstrated, motivate the development of hybrid models and ensemble models between different deep learning architectures and between deep learning models with other machine learning prediction methods in order to obtain more effective predictive models.

Thus, the main objective of our project is to develop deep learning models for temporal data in order to obtain predictive models that are more effective and efficient enough to be applied in real time. Therefore, our main objectives are, firstly, to improve efficiency through the fusion of different representations of temporal data and the fusion of models, in particular by developing hybrid models and ensemble models, and secondly, that the deep learning models we develop can be applied in real time through the design of dynamic ensembles and the improvement of efficiency through transfer learning, which will be one of the paradigms that will revolutionize the world of commercial applications in the coming years.

Aplications: Documents to be submited by aplicants: 
Applicants should send to atrolor@upo.es a research CV including all their publications, competitive research projects they have led or participated in, researchstays, felowship, grants, etc. 

 

 

 

Disclaimer:

The responsibility for the hosting offers published on this website, including the hosting description, lies entirely with the publishing institutions. The application is handled uniquely by the employer, who is also fully responsible for the recruitment and selection processes.