Scholar
rahma chaabouni
Google Scholar ID: jYN2eXEAAAAJ
DeepMind
Citations & Impact
Citations
12,691
H-index
17
i10-index
17
Publications
20
Co-authors
0
Publications
Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
- Published several papers including 'Communicating artificial neural networks develop efficient color-naming systems' (to appear in PNAS), 'What they do when in doubt: a study of inductive biases in seq2seq learners' (ICLR 2021), 'Compositionality and generalization in emergent languages' (ACL 2020), and more.
Research Experience
- Conducting PhD research at Facebook AI Research and ENS Ulm, focusing on emergent language comparisons, inductive biases studies, etc.
Education
- PhD candidate at Facebook AI Research and ENS Ulm, advised by Prof. Marco Baroni and Prof. Emmanuel Dupoux; Master of Applied Mathematics from Supelec and DataScience master’s degree from Ecole Polytechnique.
Background
- Interested in understanding what made our language unique and how we can endow artificial models with such a communication protocol. Uses deep learning techniques, building up on the emergent communication field to compare deep learner languages with human languages. Recently focusing on looking at deep learners’ inductive biases and how to inject the “right” biases in these task-agnostic models.
Co-authors