Tree-based models, particularly favored for handling tabular data (Grinsztajn et al., 2022), continue

Therefore, it's important to study whether we can and also how to

Deep learning succeeds by doing hierarchical feature learning, yet tuning hyper-parameters (HP) such as initialization scales, learning rates etc., only give indirect

In the reminder of the paper we will try whenever possible to formally introduce functions by clearly stating domain and co-domain.

We propose a novel loss function called Energy Discrepancy (ED) which does not rely on the computation of scores or expensive Markov chain Monte Carlo.