Model
AlphaTensor
An AI system introduced by DeepMind in 2022 that uses reinforcement learning to discover matrix multiplication algorithms more efficient than Strassen's algorithm.
Timeline (last 90 days)
Oct 13
Recent work such as AlphaTensor and FunSearch has used search and learning methods to discover new algorithms or new combinatorial constructions for matrix multiplication, and some results have been proven correct and are of practical value.
Unverified50%
All Facts (3)
Unverified
AlphaTensor在2022年发现了新的矩阵乘法算法
70%UnverifiedRecent work such as AlphaTensor and FunSearch has used search and learning methods to discover new algorithms or new combinatorial constructions for matrix multiplication, and some results have been proven correct and are of practical value.
50%UnverifiedDeepMind于2022年推出的AlphaTensor以强化学习方法重新发现了比Strassen算法更优的矩阵乘法路径
50%Source Articles
AI Auto-Research Breakthrough: Integer Multiplication Falls Below n log n for the First TimeOct 13Did OpenAI Solve the Wrong Navier-Stokes Problem? The Deeper Meaning of an AI Research ControversySep 22AlphaEvolve Goes Commercial: Google Cloud Opens Algorithm Discovery to Solve Enterprise Optimization ChallengesJul 11