An Explainable Probabilistic Classifier for Categorical Data Inspired to Quantum Physics

TL;DR

This paper introduces the Sparse Tensor Classifier (STC), an explainable probabilistic model inspired by quantum physics, that achieves state-of-the-art results on categorical data with minimal preprocessing.

Contribution

The paper proposes a novel quantum-inspired framework for categorical data classification, unifying classical and quantum probabilities, and providing inherent interpretability.

Findings

Achieves state-of-the-art performance on structured and text data

Requires minimal data preprocessing and hyper-parameter tuning

Provides native explanations for predictions

Abstract

This paper presents Sparse Tensor Classifier (STC), a supervised classification algorithm for categorical data inspired by the notion of superposition of states in quantum physics. By regarding an observation as a superposition of features, we introduce the concept of wave-particle duality in machine learning and propose a generalized framework that unifies the classical and the quantum probability. We show that STC possesses a wide range of desirable properties not available in most other machine learning methods but it is at the same time exceptionally easy to comprehend and use. Empirical evaluation of STC on structured data and text classification demonstrates that our methodology achieves state-of-the-art performances compared to both standard classifiers and deep learning, at the additional benefit of requiring minimal data pre-processing and hyper-parameter tuning. Moreover, STC provides a native explanation of its predictions both for single instances and for each target label globally.