Learning Kolmogorov Models for Binary Random Variables
This work aims to improve interpretable machine learning for applications like recommendation systems, though it appears incremental as it builds on existing Kolmogorov model concepts.
The paper tackles the problem of learning a Kolmogorov model for binary random variables by deriving conditions to link outcomes and extract relations from data, and proposes an algorithm with first-order optimality, applied to recommendation systems.
We summarize our recent findings, where we proposed a framework for learning a Kolmogorov model, for a collection of binary random variables. More specifically, we derive conditions that link outcomes of specific random variables, and extract valuable relations from the data. We also propose an algorithm for computing the model and show its first-order optimality, despite the combinatorial nature of the learning problem. We apply the proposed algorithm to recommendation systems, although it is applicable to other scenarios. We believe that the work is a significant step toward interpretable machine learning.