# Estimating Cardinalities with Deep Sketches

## TL;DR

Deep Sketches utilize deep learning to create compact database models that accurately estimate SQL query result sizes, capturing complex correlations and outperforming traditional estimators.

## Contribution

The paper introduces Deep Sketches, a novel deep learning approach for cardinality estimation that models correlations across columns and tables.

## Key findings

- Deep Sketches outperform traditional estimators on TPC-H and IMDb datasets.
- The approach captures complex column correlations effectively.
- Demonstration includes training, monitoring, and querying with Deep Sketches.

## Abstract

We introduce Deep Sketches, which are compact models of databases that allow us to estimate the result sizes of SQL queries. Deep Sketches are powered by a new deep learning approach to cardinality estimation that can capture correlations between columns, even across tables. Our demonstration allows users to define such sketches on the TPC-H and IMDb datasets, monitor the training process, and run ad-hoc queries against trained sketches. We also estimate query cardinalities with HyPer and PostgreSQL to visualize the gains over traditional cardinality estimators.