Towards Scalable Meta-Learning of near-optimal Interpretable Models via Synthetic Model Generations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Myint, Kyaw Hpone, Wu, Zhe, Day, Alexandre G. R., Iyengar, Giri
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911251100073984
author Myint, Kyaw Hpone
Wu, Zhe
Day, Alexandre G. R.
Iyengar, Giri
author_facet Myint, Kyaw Hpone
Wu, Zhe
Day, Alexandre G. R.
Iyengar, Giri
contents Decision trees are widely used in high-stakes fields like finance and healthcare due to their interpretability. This work introduces an efficient, scalable method for generating synthetic pre-training data to enable meta-learning of decision trees. Our approach samples near-optimal decision trees synthetically, creating large-scale, realistic datasets. Using the MetaTree transformer architecture, we demonstrate that this method achieves performance comparable to pre-training on real-world data or with computationally expensive optimal decision trees. This strategy significantly reduces computational costs, enhances data generation flexibility, and paves the way for scalable and efficient meta-learning of interpretable decision tree models.
format Preprint
id arxiv_https___arxiv_org_abs_2511_04000
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Scalable Meta-Learning of near-optimal Interpretable Models via Synthetic Model Generations
Myint, Kyaw Hpone
Wu, Zhe
Day, Alexandre G. R.
Iyengar, Giri
Machine Learning
Artificial Intelligence
Computation and Language
Decision trees are widely used in high-stakes fields like finance and healthcare due to their interpretability. This work introduces an efficient, scalable method for generating synthetic pre-training data to enable meta-learning of decision trees. Our approach samples near-optimal decision trees synthetically, creating large-scale, realistic datasets. Using the MetaTree transformer architecture, we demonstrate that this method achieves performance comparable to pre-training on real-world data or with computationally expensive optimal decision trees. This strategy significantly reduces computational costs, enhances data generation flexibility, and paves the way for scalable and efficient meta-learning of interpretable decision tree models.
title Towards Scalable Meta-Learning of near-optimal Interpretable Models via Synthetic Model Generations
topic Machine Learning
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2511.04000