Published on

Project: Data Distillation for Tabular Data

Table of contents

I had the pleasure of presenting our work with folks at IBM research at AAAI-24 as a student abstract.

This is a preliminary work, aiming to investigate data distillation applied to tabular data. Using the classification accuracy (balanced accuracy) as a metric, we show that data distillation can be used to reduce the size of the dataset while maintaining non-trivial performance on different classifiers.