Mathematical discoveries from program search with large language models
Bernardino Romera‐Paredes, Mohammadamin Barekatain, Alexander Novikov et al. · Nature · 2023
Abstract Large language models (LLMs) have demonstrated tremendous capabilities in solving complex tasks, from quantitative reasoning to understanding natural language. However, LLMs sometimes suffer from confabulations (or hallucinations), which can result in them making plausible but incorrect statements 1,2 . This hinders the use of current large models in scientific discovery. Here we introduce FunSearch (short for searching in the function space), an evolutionary procedure based on pairing a pretrained LLM with a systematic evaluator. We demonstrate the effectiveness of this approach to surpass the best-known results in important problems, pushing the boundary of existing LLM-based approaches 3 . Applying FunSearch to a central problem in extremal combinatorics—the cap set problem—we discover new constructions of large cap sets going beyond the best-known ones, both in finite dimensional and asymptotic cases. This shows that it is possible to make discoveries for established open problems using LLMs. We showcase the generality of FunSearch by applying it to an algorithmic problem, online bin packing, finding new heuristics that improve on widely used baselines. In contrast to most computer search approaches, FunSearch searches for programs that describe how to solve a problem, rather than what the solution is. Beyond being an effective and scalable strategy, discovered programs tend to be more interpretable than raw solutions, enabling feedback loops between domain experts and FunSearch, and the deployment of such programs in real-world applications.
Cited by 510 works. doi:10.1038/s41586-023-06924-6 Open-access PDF
Authors
- Bernardino Romera‐Paredes
- Mohammadamin Barekatain
- Alexander Novikov
- Matej Balog
- Manish Kumar
- Emilien Dupont
- Francisco J. R. Ruiz
- Jordan S. Ellenberg
- Pengming Wang
- Omar Fawzi
- Pushmeet Kohli
- Alhussein Fawzi
References
- A formal theory of inductive inference. Part I
- On the Length of Programs for Computing Finite Binary Sequences
- Distilling Free-Form Natural Laws from Experimental Data
- Competition-level code generation with AlphaCode
- Generative Agents: Interactive Simulacra of Human Behavior
- Approximation Algorithms for Bin-Packing — An Updated Survey
- An Introduction to Kolmogorov Complexity and Its Applications
- Additive Combinatorics
- Toolformer: Language Models Can Teach Themselves to Use Tools
- Genetic programming as a means for programming computers by natural selection