OMEGA: Optimizing Machine Learning by Evaluating Generated Algorithms

TL;DR

OMEGA is an end-to-end framework that automates AI research by generating, evaluating, and deploying novel machine learning algorithms, outperforming standard baselines on multiple datasets.

Contribution

The paper introduces OMEGA, a comprehensive system combining meta-prompt engineering and code generation to automate the creation of effective ML classifiers.

Findings

Generated algorithms outperform scikit-learn baselines on 20 benchmark datasets.

The framework successfully automates the entire process from idea to executable code.

Models are accessible via a Python package for practical use.

Abstract

In order to automate AI research we introduce a full, end-to-end framework, OMEGA: Optimizing Machine learning by Evaluating Generated Algorithms, that starts at idea generation and ends with executable code. Our system combines structured meta-prompt engineering with executable code generation to create new ML classifiers. The OMEGA framework has been utilized to generate several novel algorithms that outperform scikit-learn baselines across a robust selection of 20 benchmark datasets (infinity-bench). You can access models discussed in this paper and more in the python package: pip install omega-models.