Olivier Grisel

TabICL is a fully-open source, state of the art Tabular Foundational Model (TFM). TFMs are pretrained on large amount of synthetic data and have been advertised as being able to provide well calibrated probabilistic predictions both for classification and regression tasks.

Most common machine learning models (linear, tree-based or neural network-based), optimize for the least squares loss when trained for regression tasks. As a result, they output a point estimate of the conditional expected value of the target: E[y|X].

Machine Learning practitioners build predictive models from "noisy" data resulting in uncertain predictions. But what does "noise" mean in a machine learning context?

While the tutorial attendance is comprised in the conference pass, we ask you to register for this tutorial on https://www.jupytercon.com/tickets as the seats available are limited. -- Tutorial notebooks: - https://github.com/soda-inria/survival-analysis-benchmark According to Wikipedia:...