Algebraically Explainable Controllers: Decision Trees and Support Vector Machines Join Forces

TL;DR

This paper proposes a novel approach that combines decision trees and support vector machines to create explainable controllers capable of modeling complex algebraic relationships in continuous systems.

Contribution

The paper introduces a hybrid method that leverages the interpretability of decision trees with the modeling power of SVMs to produce understandable controllers for complex dynamics.

Findings

Effective in modeling polynomial relationships

Produces small, interpretable controllers

Validated on established benchmarks

Abstract

Recently, decision trees (DT) have been used as an explainable representation of controllers (a.k.a. strategies, policies, schedulers). Although they are often very efficient and produce small and understandable controllers for discrete systems, complex continuous dynamics still pose a challenge. In particular, when the relationships between variables take more complex forms, such as polynomials, they cannot be obtained using the available DT learning procedures. In contrast, support vector machines provide a more powerful representation, capable of discovering many such relationships, but not in an explainable form. Therefore, we suggest to combine the two frameworks in order to obtain an understandable representation over richer, domain-relevant algebraic predicates. We demonstrate and evaluate the proposed method experimentally on established benchmarks.