Machine Learning with Probabilistic Law Discovery: A Concise Introduction
It offers an interpretable alternative to methods like Decision Trees for various applications, but appears incremental as it builds on existing rule-learning approaches.
The paper introduces Probabilistic Law Discovery (PLD), a logic-based machine learning method that learns human-readable probabilistic rules for tasks like classification and anomaly detection, emphasizing transparency and interpretability.
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.