LightAutoDS-Tab: Multi-AutoML Agentic System for Tabular Data

TL;DR

LightAutoDS-Tab is a multi-AutoML system that integrates LLM-based code generation with various AutoML tools to enhance flexibility and performance on tabular data tasks, outperforming existing solutions.

Contribution

It introduces a novel multi-AutoML agentic system combining LLMs with multiple AutoML tools for improved tabular data task performance.

Findings

Outperforms state-of-the-art open-source AutoML solutions on Kaggle datasets

Enhances pipeline design flexibility and robustness

Demonstrates effectiveness across diverse tabular data tasks

Abstract

AutoML has advanced in handling complex tasks using the integration of LLMs, yet its efficiency remains limited by dependence on specific underlying tools. In this paper, we introduce LightAutoDS-Tab, a multi-AutoML agentic system for tasks with tabular data, which combines an LLM-based code generation with several AutoML tools. Our approach improves the flexibility and robustness of pipeline design, outperforming state-of-the-art open-source solutions on several data science tasks from Kaggle. The code of LightAutoDS-Tab is available in the open repository https://github.com/sb-ai-lab/LADS