This article is cited in 5 papers

Algebraic and logical methods in computer science and artificial intelligence

Machine learning with probabilistic law discovery: a concise introduction

Alexander V. Demin, Denis K. Ponomaryov

Ershov Institute of Informatics Systems SB RAS, Novosibirsk, Russian Federation

Abstract: Probabilistic Law Discovery (PLD) is a logic based Machine Learning method, which implements a variant of probabilistic rule learning. In several aspects, PLD is close to Decision Tree/Random Forest methods, but it differs significantly in how relevant rules are defined. The learning procedure of PLD solves the optimization problem related to the search for rules (called probabilistic laws), which have a minimal length and relatively high probability. At inference, ensembles of these rules are used for prediction. Probabilistic laws are human-readable and PLD based models are transparent and inherently interpretable. Applications of PLD include classification/clusterization/regression tasks, as well as time series analysis/anomaly detection and adaptive (robotic) control. In this paper, we outline the main principles of PLD, highlight its benefits and limitations and provide some application guidelines.

Keywords: probabilistic rule learning, knowledge discovery, interpretable machine learning.

UDC: 004.85

MSC: 68T05

Received: 27.12.2022
Revised: 03.02.2023
Accepted: 07.02.2023

Language: English

DOI: 10.26516/1997-7670.2023.43.91