<html><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class=""><div class=""><span style="color:rgb(0,0,0)" class="">The </span><strong style="color:rgb(0,0,0)" class="">SiMul team</strong><span style="color:rgb(0,0,0)" class=""> </span><span style="color:rgb(0,0,0)" class="">(</span><a rel="noopener" href="https://cran-simul.github.io/" target="_blank" data-saferedirecturl="https://www.google.com/url?q=https://cran-simul.github.io/&source=gmail&ust=1743195590622000&usg=AOvVaw2ey4A_HRHbWgGPWiZPQK5d" class="">https://cran-simul.<wbr class="">github.io</a><span style="color:rgb(0,0,0)" class="">) at the</span><span style="color:rgb(0,0,0)" class=""> </span><strong style="color:rgb(0,0,0)" class="">University of Lorraine</strong><span style="color:rgb(0,0,0)" class=""> </span><span style="color:rgb(0,0,0)" class="">is offering a</span><span style="color:rgb(0,0,0)" class=""> </span><strong style="color:rgb(0,0,0)" class="">fully funded PhD position</strong><span style="color:rgb(0,0,0)" class=""> </span><span style="color:rgb(0,0,0)" class="">on the</span><span style="color:rgb(0,0,0)" class=""> </span><strong style="color:rgb(0,0,0)" class="">theoretical foundations of self-supervised learning</strong><span style="color:rgb(0,0,0)" class="">, focusing on</span><span style="color:rgb(0,0,0)" class=""> </span><span style="color:rgb(0,0,0)" class="">representation stability, interpretability, and efficiency</span><span style="color:rgb(0,0,0)" class="">.</span></div><div class=""><p style="color:rgb(0,0,0)" class="">Despite
their success, self-supervised approaches and foundation models still
lack a thorough theoretical understanding. This project aims to bridge
that gap by exploring <strong class="">connections between AI models and low-rank tensor decompositions</strong>, providing a rigorous mathematical framework to address key questions:</p><ul style="color:rgb(0,0,0)" class=""><li class="">When are learned representations interpretable and stable?</li><li class="">How do models perform on heterogeneous data (e.g., federated or personalized learning)?</li><li class="">Can smaller, energy-efficient models achieve strong performance on specialized tasks?</li></ul><h3 style="color:rgb(0,0,0)" class=""><strong class=""><font size="2" class="">Position Details</font></strong></h3><ul style="color:rgb(0,0,0)" class=""><li class=""><strong class="">Location:</strong> Nancy, France</li><li class=""><strong class="">Funding:</strong> Fully funded</li><li class=""><strong class="">Candidate Profile:</strong> Master’s degree (or equivalent) in applied mathematics or an AI-related field. A strong mathematical background is required.</li><li class=""><strong class="">More details: </strong><a href="https://cran-simul.github.io/assets/jobs/sujetThese_LENTILLE_2025.pdf" target="_blank" data-saferedirecturl="https://www.google.com/url?q=https://cran-simul.github.io/assets/jobs/sujetThese_LENTILLE_2025.pdf&source=gmail&ust=1743195590622000&usg=AOvVaw1KJYWPvA_f-qCLh0va6usK" class="">https://cran-simul.<wbr class="">github.io/assets/jobs/<wbr class="">sujetThese_LENTILLE_2025.pdf</a></li></ul><h3 style="color:rgb(0,0,0)" class=""><strong class=""><font size="2" class="">How to Apply</font></strong></h3><p class=""><font color="#000000" class="">Interested candidates should send their application to </font><span style="color:rgb(0,0,0)" class="">David Brie, Ricardo Borsoi, and Konstantin Usevich </span><font color="#000000" class="">(<a href="mailto:david.brie@univ-lorraine.fr" target="_blank" class="">david.brie@univ-<wbr class="">lorraine.fr</a>, <a href="mailto:ricardo.borsoi@univ-lorraine.fr" target="_blank" class="">ricardo.borsoi@<wbr class="">univ-lorraine.fr</a>, <a href="mailto:konstantin.usevich@univ-lorraine.fr" target="_blank" class="">konstantin.<wbr class="">usevich@univ-lorraine.fr</a>) </font><span style="color:rgb(0,0,0)" class="">with<wbr class="">:</span></p><ul style="color:rgb(0,0,0)" class=""><li class="">An academic CV</li><li class="">A short explanation of research interests and motivation for this position</li></ul></div></body></html>