Vivien Cabannes

Structured Prediction with Partial Labelling through the Infimum Loss

2020 · arXiv (Cornell University)

Annotating datasets is one of the main costs in nowadays supervised learning. The goal of weak supervision is to enable models to learn using only forms of labelling which are cheaper to collect, as partial …

Super fast rates in structured prediction

2021 · arXiv (Cornell University)

Discrete supervised learning problems such as classification are often tackled by introducing a continuous surrogate problem akin to regression. Bounding the original error, between estimate and solution, by the surrogate error endows discrete problems with …

A Mechanistic Study of Transformers Training Dynamics

2024 · arXiv (Cornell University)

Large-scale pretraining of transformers has been central to the success of foundation models. However, the scale of those models limits our understanding of the mechanisms at play during optimization. In this work, we study the …