Interactive Evolution

Interactive evolution integrates human feedback into evolutionary algorithms to guide the search for optimal solutions across diverse domains, from software architecture design to artistic image generation and even the simulation of complex biological systems. Current research focuses on improving efficiency and user experience through techniques like regret-based elicitation, adaptive sampling within behavioral spaces (e.g., using Quality Diversity algorithms), and the incorporation of large language models to manage complex design spaces and facilitate human-computer collaboration. This approach promises to enhance the effectiveness of evolutionary computation by leveraging human expertise and intuition, leading to more efficient and user-satisfying solutions in various scientific and engineering applications.