Post-Doctoral Research Visit F/M Event-based unsupervised waveform learning for physiological signals

Le descriptif de l’offre ci-dessous est en Anglais

Type de contrat : CDD

Niveau de diplôme exigé : Thèse ou équivalent

Fonction : Post-Doctorant

A propos du centre ou de la direction fonctionnelle

The Inria Saclay Research Centre was established in 2008. It has developed as part of the Saclay site in partnership with Paris-Saclay University and with the Institut Polytechnique de Paris since 2021.

The centre has 39 project teams , 27 of which operate jointly with Paris-Saclay University and the Institut Polytechnique de Paris. Its activities occupy over 600 scientists and research and innovation support staff, including 54 different nationalities.

Contexte et atouts du poste

The postdoc will take place in Inria Saclay, in the MIND team. This is a large team working focused on mathematical methods for statistical modeling of brain function using neuroimaging data (fMRI, MEG, EEG). Particular topics of interest include machine learning techniques, numerical and parallel optimization, applications to human cognitive neuroscience, event detection, and scientific software development. A particular emphasis is put on interdisciplinary projects.

Mission confiée

A natural way to describe physiological signals that is compatible with events is to consider recurring patterns, whose localization can be seen as a train of events. These descriptions are already used by both learning-based methods and in more manual pipelines. For instance, in neuroscience, the interest in transient waveforms to describe in M/EEG recordings has risen in recent years as markers of cognitive functions or pathologies. The success of these representations mostly depends on the way to select the patterns. However, most methods consider that the occurrences of physiological events are independent and cannot incorporate knowledge from external events. This leads to unreliable event extraction, where spurious and non-plausible events are detected. Moreover, this makes it harder to highlight global properties in the signal, such as the rhythms or the link between different events. The goal of this postdoc will be to develop efficient end-to-end procedures to detect and model events in physiological signals, accounting for their inter-dependence patterns. In particular, we will aim to extend Convolutional Dictionary Learning (CDL) to the case where its activations follow PP models, which describe both the activations and external events. The major challenge will be to propose efficient and reliable solvers to solve the resulting optimization problem. Due to the scale of the problem (commonly over 100,000 time points), this step will require large-scale and distributed optimization, which will benefit from previous work on distributed solvers for CDL in the team.

Principales activités

main activities :

- Read papers and state of the art
- Benchmark existing algorithms
- Adapt the formulation to the target scenario.
- Program, run, and analyze simulation results.

Complementary activities

- Participate in the team's activities: scientific meetings, seminars, and scientific presentations.

Compétences

  • Strong mathematical background. Knowledge in machine learning is a plus.

  • Good programming skills in Python. Knowledge of a deep learning framework is a plus.

  • The candidate should be proficient in English. Knowing French is not necessary, as daily communication in the team is mostly in English due to the strong international environment.

Avantages

  • Subsidized meals
  • Partial reimbursement of public transport costs
  • Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
  • Possibility of teleworking and flexible organization of working hours
  • Professional equipment available (videoconferencing, loan of computer equipment, etc.)
  • Social, cultural and sports events and activities

Rémunération

2788 € gross/month