Google's artificial intelligence (AI) research lab DeepMind has unveiled an advanced agent, AlphaEvolve, which can target fundamental and complex mathematics and computing problems.
It has the versatility of large language models (LLMs), which can summarise documents, generate code, and generate new ideas. It also goes a step ahead by verifying answers through automated evaluators.
One of the major threats facing the nascent AI world is hallucinations by chatbots. What AlphaEvolve does is that it uses LLMs to generate answers to prompts, and automatically evaluates and scores these answers for accuracy.
Researchers have used this technique before, but according to DeepMind, the 'state-of-the-art' Gemini Flash and Gemini Pro models make AlphaEvolve more capable. "Together, these models propose computer programs that implement algorithmic solutions as code," DeepMind said in a blog post.
Google also deployed AlphaEvolve on its own infrastructure to test it across practical problems. As per the blog post, AlphaEvolve enhanced the efficiency of Google's data centres, chip design and AI training processes, including training the large language models underlying AlphaEvolve itself.
"By finding smarter ways to divide a large matrix multiplication operation into more manageable subproblems, it sped up this vital kernel in Gemini’s architecture by 23%, leading to a 1% reduction in Gemini's training time," the lab said.
To test AlphaEvolve’s breadth, DeepMind applied the system to over 50 open problems in mathematical analysis, geometry, combinatorics and number theory. In roughly 75% of cases, AlphaEvolve "rediscovered" best-known solutions, and in 20% of the cases, it improved upon these solutions.
It has the versatility of large language models (LLMs), which can summarise documents, generate code, and generate new ideas. It also goes a step ahead by verifying answers through automated evaluators.
One of the major threats facing the nascent AI world is hallucinations by chatbots. What AlphaEvolve does is that it uses LLMs to generate answers to prompts, and automatically evaluates and scores these answers for accuracy.
Researchers have used this technique before, but according to DeepMind, the 'state-of-the-art' Gemini Flash and Gemini Pro models make AlphaEvolve more capable. "Together, these models propose computer programs that implement algorithmic solutions as code," DeepMind said in a blog post.
Google also deployed AlphaEvolve on its own infrastructure to test it across practical problems. As per the blog post, AlphaEvolve enhanced the efficiency of Google's data centres, chip design and AI training processes, including training the large language models underlying AlphaEvolve itself.
"By finding smarter ways to divide a large matrix multiplication operation into more manageable subproblems, it sped up this vital kernel in Gemini’s architecture by 23%, leading to a 1% reduction in Gemini's training time," the lab said.
To test AlphaEvolve’s breadth, DeepMind applied the system to over 50 open problems in mathematical analysis, geometry, combinatorics and number theory. In roughly 75% of cases, AlphaEvolve "rediscovered" best-known solutions, and in 20% of the cases, it improved upon these solutions.
