# AgentGA: Evolving Code Solutions in Agent-Seed Space

arXiv: https://arxiv.org/abs/2604.14655 · HTML: https://mlsanity.com/paper/2604.14655/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 33 references that are in the map.
- First posted on arXiv 16 Apr 2026.

## Builds on (most cited first, 23 of 23)

- [Competition-Level Code Generation with AlphaCode](https://mlsanity.com/paper/2203.07814/) (2022)
- [AlphaEvolve: A coding agent for scientific and algorithmic discovery](https://mlsanity.com/paper/2506.13131/) (2025)
- [AutoGluon-Tabular: Robust and Accurate AutoML for Structured Data](https://mlsanity.com/paper/2003.06505/) (2020)
- [Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents](https://mlsanity.com/paper/2505.22954/) (2025)
- [ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolution](https://mlsanity.com/paper/2402.01145/) (2024)
- [AIDE: AI-Driven Exploration in the Space of Code](https://mlsanity.com/paper/2502.13138/) (2025)
- [Data Interpreter: An LLM Agent For Data Science](https://mlsanity.com/paper/2402.18679/) (2024)
- [EvoPrompt: Connecting LLMs with Evolutionary Algorithms Yields Powerful Prompt Optimizers](https://mlsanity.com/paper/2309.08532/) (2023)
- [Auto-Sklearn 2.0: Hands-free AutoML via Meta-Learning](https://mlsanity.com/paper/2007.04074/) (2020)
- [AgentSquare: Automatic LLM Agent Search in Modular Design Space](https://mlsanity.com/paper/2410.06153/) (2024)
- [Large Language Models for Automated Data Science: Introducing CAAFE for Context-Aware Automated Feature Engineering](https://mlsanity.com/paper/2305.03403/) (2023)
- [DS-Agent: Automated Data Science by Empowering Large Language Models with Case-Based Reasoning](https://mlsanity.com/paper/2402.17453/) (2024)
- [AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML](https://mlsanity.com/paper/2410.02958/) (2024)
- [Evaluation of a Tree-based Pipeline Optimization Tool for Automating Data Science](https://mlsanity.com/paper/1603.06212/) (2016)
- [AutoKaggle: A Multi-Agent Framework for Autonomous Data Science Competitions](https://mlsanity.com/paper/2410.20424/) (2024)
- [SE-Agent: Self-Evolution Trajectory Optimization in Multi-Step Reasoning with LLM-Based Agents](https://mlsanity.com/paper/2508.02085/) (2025)
- [MLCopilot: Unleashing the Power of Large Language Models in Solving Machine Learning Tasks](https://mlsanity.com/paper/2304.14979/) (2023)
- [EvoFlow: Evolving Diverse Agentic Workflows On The Fly](https://mlsanity.com/paper/2502.07373/) (2025)
- [Group-Evolving Agents: Open-Ended Self-Improvement via Experience Sharing](https://mlsanity.com/paper/2602.04837/) (2026)
- [Kolb-Based Experiential Learning for Generalist Agents with Human-Level Kaggle Data Science Performance](https://mlsanity.com/paper/2411.03562/) (2024)
- [SELA: Tree-Search Enhanced LLM Agents for Automated Machine Learning](https://mlsanity.com/paper/2410.17238/) (2024)
- [MLZero: A Multi-Agent System for End-to-end Machine Learning Automation](https://mlsanity.com/paper/2505.13941/) (2025)
- [I-MCTS: Enhancing Agentic AutoML via Introspective Monte Carlo Tree Search](https://mlsanity.com/paper/2502.14693/) (2025)

## Built on by (most cited first, 1 of 1)

- [Escher-Loop: Mutual Evolution by Closed-Loop Self-Referential Optimization](https://mlsanity.com/paper/2604.23472/) (2026)

## Abstract (arXiv)

We present AgentGA, a framework that evolves autonomous code-generation runs by optimizing the agent seed: the task prompt plus optional parent archives that initialize a fresh workspace. The outer loop searches over these reusable starting conditions rather than editing code directly. Each generation launches a fresh autonomous run in an isolated workspace, while selected parent archives provide inherited artifacts that descendants can inspect and reuse. AgentGA couples a population-level genetic algorithm with long-horizon agents; selection uses deterministic 1:1 elite tournaments and operator allocation is adapted online with a modified Hedge controller. We instantiate the approach for tabular AutoML on the 16-competition Weco-Kaggle Lite benchmark. Across the full benchmark, AgentGA averages 71.90% Exceeds % of Human versus 51.38% for the AIDE reference, winning 15/16 competitions. Within AgentGA runs, descendants conditioned on inherited parent archives win 51.9% of 1,680 parent-child tournaments versus 8.6% for de novo proposals. These results support agent-seed optimization as a practical design choice for autonomous code-search systems.