PhD Supervisor: David Camacho (UPM); Auxiliary supervisors: Salvatore Cuomo (UNINA), Dariusz Mrozek (SUT), Jia-Chun Lin (NTNU); R&D cooperation: GMV

Objectives: The adoption of deep learning has been slower for time series than for computer vision and natural language processing. The emergence of techniques that are based on deep learning such as generative AI, transfer learning (which involves taking a pre-trained model from one task and fine-tuning it on a different but related task), the use of pre-trained models (ML models that are trained on a diverse range of time series data and can be used as feature extractors or for transfer learning), self-supervised learning (a technique where the model generates its own labels from the data, thereby making it particularly useful when labelled data is scarce or expensive to obtain), provide a new environment that would enable exploring and discovering innovative methods for time series analysis and prediction at large scales.

Expected Results: The main result of this project will be the development and implementation of state-of-the-art deep learning pipelines for time series data, which will scale up the vast amount of time series data that is available in a wide range of domains. The project will be used in various industrial domains, e.g., space or industry in order to demonstrate the feasibility of the implemented models.

Applied research: The primary objective of DC2 is to design and implement new and deep learning-based time series data techniques, the project will be applied to  industrial sectors (e.g. space under the collaboration with GMV), to validate the practicality and feasibility of the new time series models.

Planned secondments: UNINA (4 months); SUT(4 months);NTNU(4 months)                    

Enrolment in Doctoral degree: UPM

Recruitment of Doctoral Candidate: https://euraxess.ec.europa.eu/jobs/288653