Fast Private Parameter Learning and Inference for Sum-Product Networks

TL;DR

This paper presents a privacy-preserving method for learning the weights of sum-product networks using secret sharing, enabling efficient and secure parameter learning and inference in distributed settings.

Contribution

It introduces a novel secret sharing-based approach for private weight learning in SPNs with fixed architecture, improving speed and resource efficiency.

Findings

Secure weight learning with secret sharing is feasible.

The method achieves fast computations with low resource requirements.

Private inference on learned SPNs is effectively demonstrated.

Abstract

A sum-product network (SPN) is a graphical model that allows several types of inferences to be drawn efficiently. There are two types of learning for SPNs: Learning the architecture of the model, and learning the parameters. In this paper, we tackle the second problem: We show how to learn the weights for the sum nodes, assuming the architecture is fixed, and the data is horizontally partitioned between multiple parties. The computations will preserve the privacy of each participant. Furthermore, we will use secret sharing instead of (homomorphic) encryption, which allows fast computations and requires little computational resources. To this end, we use a novel integer division to compute approximate real divisions. We also show how simple and private inferences can be performed using the learned SPN.