Cognitive neuroscience
Ecole Normale Supérieure – PSL / Inserm
valentin.wyart[at]ens.fr
Valentin leads a research group at the Computational Cognitive Neuroscience Lab of the École Normale Supérieure since 2018. His research group studies learning and decision-making under uncertainty in humans and artificial cognitive systems, for which he was awarded a starting grant from the European Research Council.
Learning and decision-making under uncertainty in humans and artificial cognitive systems.
Astrophysics
Observatoire de Paris – LERMA, Univ. PSL
david.cornu[at]observatoiredeparis.psl.eu
French numerical astrophysicist, working at the Paris Observatory as a member of the LERMA. His current research topic revolves around developing state-of-the-art Machine Learning approaches and tools for analyzing massive datasets from modern giant interferometers (LOFAR, ALMA, NenuFAR, MeerKAT, SKA, …) in the context of the MINERVA project.
Galaxy detection and characterization in radio-astronomical datasets. MINERVA project (MachINe lEarning for Radioastronomy at obserVatoire de Paris).
Vincent Divol is an AI Junior Fellow at PSL. His research interests broadly lie at the intersection of statistics, machine learning and geometry.
Statistical optimal transport
Computational Neuroscience
Laboratoire de Neurosciences Cognitives Computationnelles, Ecole Normale Superieure
srdjan.ostojic[at]ens.psl.eu
Interpretable network models of neural computations
Master degree in Systems Biology at Sorbonne University
Study of the neural correlates of social skills.
My work aims to better understand the development of social skills. In this vein, I use hyperscanning techniques, which allow the recording of the brain activity from at least two individuals engaged in social exchange, providing a novel type of neural correlate: the inter-brain synchrony. This study contributes to a deeper understanding of how the social brain develops and how rapport management occurs by looking at moments of high or low rapport with neural synchrony.
Postdoctoral researcher
École Normale Supérieure - PSL
alex.barbier-chebbah [at] pasteur.fr
PHD at Sorbonne University
Multi-Armed Bandit model.
My main research interests are random walks theory and sequential learning, focusing on their connections to decision-making task in complex environment. We combine statistical physics, Bayesian inference, information theory and numerical simulation to both probe learning procedure in insect behavior, but also to design lightweight algorithms able to mimic such procedures. In particular, relying on infotaxis methods, we develop a new class of multi-armed bandit (MAB) algorithms to achieve optimal performance at all timescales of the sequential learning procedure.
PhD student
Inria
biswesh.mohapatra [at] inria.fr
Integrated Master of Technology in Computer Science Engineering from IIIT Bangalore
Improving multimodal dialogue systems through conversational grounding.
This project plans to dive deep into the issues regarding conversational grounding. The thesis intends to do the following – 1) It will investigate why modern neural networks trained on vast amounts of data are unable to solve the phenomenon of conversational grounding in current dialogue systems. 2) It will investigate old approaches to conversational grounding for neural network-based models. 3) It will look into the role of nonverbal behavior such as eye gaze and head nods in conversational grounding, and how insights from cognitive science studies of these phenomena can be integrated into deep learning approaches. 4) The project aims to then build computational models that take conversational grounding into account and help the state-of-the-art conversational models like BlendorBot [5] or dialoGPT [6] to generate more consistent dialogues. 5) It will also develop methods to quantify and test the phenomenon of conversational grounding. 6) Finally, it will evaluate the success of these models in human-chatbot conversation, by looking at whether users are more successful in human-computer collaborative tasks.
Machine learning and statistical physics
École Normale Supérieure - PSL
tony.bonnaire [at] ens.fr
PhD in Astrophysics at Université Paris-Saclay
Machine learning and statistical physics.
My current research focuses on understanding the dynamics of simple neural networks and particularly how gradient descent can achieve good generalization, especially when initialized randomly, in high-dimensional, rough and non-convex landscapes. For this purpose, I use methods coming from theoretical physics, and more precisely the statistical physics of disordered systems, to obtain asymptotic success conditions for these methods but also to study the topological properties of the random landscapes.
Engineer
ICM Institute
camille.brianceau [at] icm-institute.org
ClinicaDL
ClinicaDL is an open-source software for deep learning processing on neuro-imaging data. My works consists in extending this software with new features and standard deep learning tools of the community, and providing PhD students and researchers with support.
Institut Pasteur
robin.cremese [at] pasteur.fr
Master’s degree – INSA Toulouse
Visualization and live segmentation from medical CT scans and MRIs: a new approach mixing probabilistic learning and Virtual reality.
Development of learning methods for optimal path planning in a constrained environment. Reinforcement Learning methods as well as physical simulations will be studied. Possible applications to the medical domain for the preventive planning of operations.
Robust Neural Machine Translation.
Neural machine translation models struggle to translate texts that differ from the “standard” data commonly used to train them. In particular, social media texts pose many challenges because they tend to be “noisy”: non-standard use of spelling, grammar and vocabulary; typographical errors; use of emojis, hashtags and at-mentions; etc. I aim to develop new methods to better translate these texts.
Phd student
Université Paris Cité and Kap Code
emmalepriol [at] gmail.com
Master’s degree : mathematics (Université Paris – Dauphine) and social sciences (Sciences Po Paris)
Using NLP to leverage social media data in the study of rare diseases.
My PhD thesis aims at exploring NLP techniques to study the online contents from rare diseases’ patients or their caregivers. The first goal is to better understand natural histories of the studied diseases, and compare the spontaneously reported symptoms to symptoms collected during medical interviews. The other goal is to study how patients’ associations become invested in public policy governance, in particular by acquiring a vast knowledge collectively.
PhD student
Inria
jakob.maier {at} inria.fr
Bachelor of Science in Mathematics: Technical University of Munich
Master of Science in Statistics and Probability: Ecole Polytechnique (M1) and Université Paris Saclay (M2)
Efficient algorithms for information extraction on graphs.
We examine algorithms that extract information from a given graph which can be issued from an application or from random sampling. The information is typically obtained in the form of statistical, algebraic, or combinatorial invariants and serves in several applications: detection of a latent geometric structure, alignment of two graphs, or community identification. The main objective is to obtain theoretical guarantees for the functioning of the algorithms while assuring that they can be efficiently executed.
PhD student
Inria
matthieu.blanke {at} inria.fr
Master M2 Mathématiques, Vision, Apprentissage of ENS Paris-Saclay in 2021 and ingénieur de l’École polytechnique.
Deep implicit layers with applications to physical systems and reinforcement learning.
My research focuses on the use of machine learning tools to understand and control physical systems. I am interested in developing efficient methods to learn physical systems, with the concern of incorporating and taking advantage of available prior information.
PhD student
Inria
ricardo-jose.garcia-pinel [at] inria.fr
Learning visuomotor policies for robotic manipulation and navigation.
Current industrial robots are mostly restricted to predefined tasks in controlled environments and are lacking the ability to adapt to new settings. Hence, the use of robots in cluttered and changing environments where the robot is required to adapt behaviors during task execution presents a major challenge. Existing methods rely on hand-engineered methods which typically overfit to a particular setting and need to be re-designed for every new task and environment. To deal with those challenges, future robots should be equipped with advanced perception coupled with control enabling navigation and manipulation in previously unseen and dynamically changing environments. The scientific objective of this thesis is to design, develop and evaluate new models and algorithms for learning visuomotor policies for goalconditioned robotic manipulation and navigation.
Physics and Learning for Visually-Guided Robotics.
The purpose of this PhD is to enhance robots autonomy in unknown environments by learning visually-guided behaviors from sensory data. We plan to automate action plans by developing methods combining modern computer vision techniques with geometric reasoning which will lead to fast search methods, able to adapt on-the-fly to previously unseen environments.
PhD student
Inria
gaspard.beugnot [at] inria.fr
Ecole Polytechnique – MVA
Non-convex optimization and learning theory with kernel methods.
Kernel methods are a versatile tool to study the statistical properties of a vast category of learning algorithm. On one hand, we aim at understanding the generalisation properties of neural network. This enable to design new and more efficient learning routines. On the other, we tackle non-convex optimisation problems through kernel sum-of-squares.
Masters of Engineering, Ecole des Ponts
Cheap and expressive neural contextual representations for textual data.
Neural language models are pre-trained using self-supervised learning to produce contextual representations of text data like words or sentences. These representations are shaped by the pre-training procedure: its data, its task, its optimization scheme, among others. I aim at identifying ways to improve the quality of the text representations by leveraging new pre-training approaches, in order to reduce data and/or compute requirements without quality loss.
PhD student
Institut Jean Nicod, ENS-PSL & Chargée d’études CNRS
aidaelamrani [at] outlook.fr
Master in Theoretical Computer Science / Aix-Marseille Université
Information in the Interplay Between Mind and Matter.
Ingénieur Civil des Mines de Paris – Mines ParisTech (Master’s degree in Science and Executive Engineering)
Master 2 – Mathématiques, Vision & Apprentissage – ENS Paris-Saclay (MVA Master’s degree)
Spatial and Temporal Heterogeneity of single cell transcriptomic data.
Spatial transcriptomics is a new kind of technology that allows biologists to measure both transcriptomic information and spatial locations in tissues. My project aims to develop methods (such as clustering, dimensionality reduction algorithm…) for the analysis of such multimodal data. In particular, we are mainly interested in the links between spatial organization and transcriptomic heterogeneity within tissues.