[D] Deep Learning VS XGBoost for tabular data: a quick test

Once per year, I write a post here on Reddit about our projects on deep learning for tabular data, and I hope this year will be no exception 🙂 Meanwhile, I have shared some results where we compare models from our previous papers with XGBoost on the datasets from the recent paper "Why do tree-based models still outperform deep learning on typical tabular data?". For us, this benchmark is a new one, so it was really interesting to check whether …

data datasets deep learning exception machinelearning per projects reddit tabular tabular data test xgboost