Reinforced Decision Trees

TL;DR

Reinforced Decision Trees are a novel model that jointly learns the category hierarchy and classification function using reinforcement learning, enabling efficient classification with low inference complexity in a single training step.

Contribution

This paper introduces Reinforced Decision Trees, integrating hierarchy learning and classification into one reinforcement learning-based algorithm, unlike traditional methods that separate these steps.

Findings

Achieves low inference complexity similar to existing hierarchical models.

Simultaneously learns category structure and classifier in one training process.

Demonstrates effectiveness on large-scale classification tasks.

Abstract

In order to speed-up classification models when facing a large number of categories, one usual approach consists in organizing the categories in a particular structure, this structure being then used as a way to speed-up the prediction computation. This is for example the case when using error-correcting codes or even hierarchies of categories. But in the majority of approaches, this structure is chosen \textit{by hand}, or during a preliminary step, and not integrated in the learning process. We propose a new model called Reinforced Decision Tree which simultaneously learns how to organize categories in a tree structure and how to classify any input based on this structure. This approach keeps the advantages of existing techniques (low inference complexity) but allows one to build efficient classifiers in one learning step. The learning algorithm is inspired by reinforcement learning and policy-gradient techniques which allows us to integrate the two steps (building the tree, and learning the classifier) in one single algorithm.