Automated Design of Agentic Systems
Shengran Hu, Cong Lu, Jeff Clune
2024-08-19
Summary
This paper introduces a new area of research called Automated Design of Agentic Systems (ADAS), which focuses on creating intelligent systems that can design and improve themselves automatically.
What's the problem?
Developing powerful general-purpose agents, which are systems that can perform a variety of tasks, is currently very labor-intensive and often relies on human-designed solutions. This process can be inefficient and may not fully utilize the potential of machine learning.
What's the solution?
The authors propose using a new approach where agents can be defined in code, allowing a 'meta agent' to program and discover new agents automatically. This method takes advantage of programming languages' capabilities to create any kind of agentic system, including new tools and workflows. They introduce an algorithm called Meta Agent Search, which helps the meta agent iteratively create and improve agents based on previous discoveries.
Why it matters?
This research is important because it could lead to the development of more advanced and flexible AI systems that can adapt and improve over time without needing constant human intervention. By automating the design process, we could create smarter agents that better serve various needs in fields like robotics, software development, and beyond.
Abstract
Researchers are investing substantial effort in developing powerful general-purpose agents, wherein Foundation Models are used as modules within agentic systems (e.g. Chain-of-Thought, Self-Reflection, Toolformer). However, the history of machine learning teaches us that hand-designed solutions are eventually replaced by learned solutions. We formulate a new research area, Automated Design of Agentic Systems (ADAS), which aims to automatically create powerful agentic system designs, including inventing novel building blocks and/or combining them in new ways. We further demonstrate that there is an unexplored yet promising approach within ADAS where agents can be defined in code and new agents can be automatically discovered by a meta agent programming ever better ones in code. Given that programming languages are Turing Complete, this approach theoretically enables the learning of any possible agentic system: including novel prompts, tool use, control flows, and combinations thereof. We present a simple yet effective algorithm named Meta Agent Search to demonstrate this idea, where a meta agent iteratively programs interesting new agents based on an ever-growing archive of previous discoveries. Through extensive experiments across multiple domains including coding, science, and math, we show that our algorithm can progressively invent agents with novel designs that greatly outperform state-of-the-art hand-designed agents. Importantly, we consistently observe the surprising result that agents invented by Meta Agent Search maintain superior performance even when transferred across domains and models, demonstrating their robustness and generality. Provided we develop it safely, our work illustrates the potential of an exciting new research direction toward automatically designing ever-more powerful agentic systems to benefit humanity.