Soufiane Belharbi bio photo

Soufiane Belharbi

Interested in Machine Learning, Deep Learning. I like running uphills (since childhood), and also running for long distance. Current challenge: break my 10km running record 00:46min:30.90 (June.02.2018 at Rouen) [STRYD].

Twitter   G. Scholar LinkedIn Github

Hi!

My name is Soufiane Belharbi. I am currently a post-doctoral research fellow at le Laboratoire d’imagerie, de vision et d’intelligence artificielle (LIVIA lab) at l’École de technologie supérieure (ÉTS) in collaboration with McGill university’s Rosalind & Morris Goodman Cancer Research Centre (GCRC). I am working with Profs. Eric Granger, Ismail Ben Ayed, and Luke McCaffrey on deep learning models for weakly supervised learning scenarios (semi-supervised learning, multi-instance learning, transfer learning) applied to medical imaging.

I completed my PhD in computer science at the Institut National des Sciences Appliquées de Rouen Normandie (INSA Rouen Normandie) in LITIS lab, Apprentissage (Learning) team (Oct. 2014-Jul.2018). I was supervised by Prof. Sébastien Adam (director), Clément Chatelain, and Romain Hérault. During my PhD, I conducted research on the regularization of neural networks through representation learning with particular focus on learning scenarios where only few training samples are available. [BibTeX] [ArXiv][HAL-Thèse][Presentation]

@phdthesis{sbelharbiphd2018,
  author = {Soufiane Belharbi},
  title = {Neural Networks Regularization Through Representation Learning},
  school = {Normandie Université, INSA Rouen Normandie, LITIS laboratory},
  year = {2018},
  pages = {196},
  note = {Supervisor: Sébastien Adam, Advisors: Clément Chatelain, Romain Hérault},
  url = {https://arxiv.org/abs/1807.05292}
}

I am intrigued and particularly interested in representation learning in neural networks.

Research Interests

  • Machine Learning
  • Deep Learning
  • Representation Learning
  • Regularization
  • Weakly supervised learning: semi-supervised/multi-instance/transfer learning

News

@phdthesis{sbelharbiphd2018,
  author = {Soufiane Belharbi},
  title = {Neural Networks Regularization Through Representation Learning},
  school = {Normandie Université, INSA Rouen Normandie, LITIS laboratory},
  year = {2018},
  pages = {196},
  note = {Supervisor: Sébastien Adam, Advisors: Clément Chatelain, Romain Hérault},
  url = {https://arxiv.org/abs/1807.05292}
}