# Batch Tournament Selection for Genetic Programming

## TL;DR

This paper introduces Batch Tournament Selection (BTS), a faster hybrid method combining tournament and lexicase selection, achieving similar solution quality with significantly reduced computational cost in genetic programming.

## Contribution

The paper presents BTS, a novel selection method that approximates lexicase selection's quality while being approximately ten times faster, offering a new efficient approach for genetic programming.

## Key findings

- BTS is up to 25 times faster than lexicase selection.
- BTS achieves comparable mean absolute error to lexicase selection.
- BTS and lexicase selection show almost no difference in diversity and performance.

## Abstract

Lexicase selection achieves very good solution quality by introducing ordered test cases. However, the computational complexity of lexicase selection can prohibit its use in many applications. In this paper, we introduce Batch Tournament Selection (BTS), a hybrid of tournament and lexicase selection which is approximately one order of magnitude faster than lexicase selection while achieving a competitive quality of solutions. Tests on a number of regression datasets show that BTS compares well with lexicase selection in terms of mean absolute error while having a speed-up of up to 25 times. Surprisingly, BTS and lexicase selection have almost no difference in both diversity and performance. This reveals that batches and ordered test cases are completely different mechanisms which share the same general principle fostering the specialization of individuals. This work introduces an efficient algorithm that sheds light onto the main principles behind the success of lexicase, potentially opening up a new range of possibilities for algorithms to come.