AlphaEvolve is a generic evolutionary coding agent that combines the generative capabilities of LLMs with automated evaluation in an iterative evolutionary framework that proposes, tests, and refines algorithmic solutions to challenging scientific and practical problems. In this paper we showcase AlphaEvolve as a tool for autonomously discovering novel mathematical constructions and advancing our understanding of long-standing open problems. To demonstrate its breadth, we considered a list of 67 problems spanning mathematical analysis, combinatorics, geometry, and number theory. The system rediscovered the best known solutions in most of the cases and discovered improved solutions in several. In some instances, AlphaEvolve is also able to generalize results for a finite number of input values into a formula valid for all input values. Furthermore, we are able to combine this methodology with Deep Think and AlphaProof in a broader framework where the additional proof-assistants and reasoning systems provide automated proof generation and further mathematical insights. These results demonstrate that large language model-guided evolutionary search can autonomously discover mathematical constructions that complement human intuition, at times matching or even improving the best known results, highlighting the potential for significant new ways of interaction between mathematicians and AI systems. We present AlphaEvolve as a powerful new tool for mathematical discovery, capable of exploring vast search spaces to solve complex optimization problems at scale, often with significantly reduced requirements on preparation and computation time.

A team of researchers and engineers has been secretly working for three years on one of humanity’s most devilish enigmas, the solution of which is considered imminent thanks to artificial intelligence

Origin

New AI agent evolves algorithms for math and practical applications in computing by combining the creativity of large language models with automated evaluators

Wir reden über eine interessante Naturkonstante, den Exponenten der Matrixmultiplikation. Ist er zwei, wäre Matrixmultiplikation genauso schnell möglich wie Matrixaddition. Ist er nicht zwei, was soll...

A discussion, prompted by Google's recent AlphaEvolve announcement, of the dangers of unsupervised, self-directed AI learning.

Discover how Google’s AlphaEvolve AI agent reclaimed 0.7% of global compute capacity, shattered matrix multiplication records, and set a new standard for production-grade autonomous systems.

Posted by Fysi with a score of 360

“History is not just the evolution of technology; it is the evolution of thought.”

New AI agent evolves algorithms for math and practical applications in computing by combining the creativity of large language models with automated evaluators