MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery
Researchers have developed MLEvolve, a novel LLM-based multi-agent framework designed for automated machine learning algorithm discovery. This framework improves upon existing methods by addressing information isolation, memory limitations, and hierarchical control issues. MLEvolve utilizes advanced search mechanisms, a retrospective memory system, and adaptive coding strategies to achieve state-of-the-art performance on benchmark tasks, even outperforming specialized methods. AI
IMPACT This framework could accelerate the development and discovery of new machine learning algorithms by automating a complex and time-consuming process.