The Duck's Brain: Training and Inference of Neural Networks in Modern Database Engines

TL;DR

This paper demonstrates how modern database systems can effectively perform neural network training and inference using relational algebra and array data types, highlighting their suitability for matrix operations despite some performance trade-offs.

Contribution

It introduces methods to implement neural network training and inference directly in SQL using relational and array data types, bridging database systems and machine learning workflows.

Findings

Array data types outperform relational representations in runtime and memory.

Modern databases efficiently handle matrix algebra for neural networks.

Relational algebra can support machine learning tasks with appropriate data transformations.

Abstract

Although database systems perform well in data access and manipulation, their relational model hinders data scientists from formulating machine learning algorithms in SQL. Nevertheless, we argue that modern database systems perform well for machine learning algorithms expressed in relational algebra. To overcome the barrier of the relational model, this paper shows how to transform data into a relational representation for training neural networks in SQL: We first describe building blocks for data transformation, model training and inference in SQL-92 and their counterparts using an extended array data type. Then, we compare the implementation for model training and inference using array data types to the one using a relational representation in SQL-92 only. The evaluation in terms of runtime and memory consumption proves the suitability of modern database systems for matrix algebra, although specialised array data types perform better than matrices in relational representation.