Transition Matrix Representation of Trees with Transposed Convolutions

TL;DR

This paper introduces TART, a novel tree representation using transposed convolutions that enhances the efficiency and accuracy of structural search in interpretable tree models.

Contribution

TART provides a new generalized tree representation that improves search efficiency and classification accuracy without relying on transition matrices.

Findings

TART achieves higher accuracy than baseline models.

It speeds up inference by avoiding transition matrix creation.

It simplifies the structural search process.

Abstract

How can we effectively find the best structures in tree models? Tree models have been favored over complex black box models in domains where interpretability is crucial for making irreversible decisions. However, searching for a tree structure that gives the best balance between the performance and the interpretability remains a challenging task. In this paper, we propose TART (Transition Matrix Representation with Transposed Convolutions), our novel generalized tree representation for optimal structural search. TART represents a tree model with a series of transposed convolutions that boost the speed of inference by avoiding the creation of transition matrices. As a result, TART allows one to search for the best tree structure with a few design parameters, achieving higher classification accuracy than those of baseline models in feature-based datasets.