Mathematicians solve decades-old mystery about the hidden order in high-dimensional randomness
Phys.org • • general
Key Points:
- Three mathematicians—Dongming Hua, Antoine Song, and Stefan Tudose—have proven Talagrand's convexity conjecture, a long-standing problem in mathematics posed in 1995 by Abel Prize winner Michel Talagrand.
- The conjecture asked whether convexity can be created in a fixed number of Minkowski sum operations in any dimension, a problem complicated by the exponential growth of geometric complexity in higher dimensions.
- By reformulating the problem in terms of probability theory, the team showed that any 1-subgaussian random vector in n dimensions can be expressed as the sum of three standard Gaussian vectors, thus solving the conjecture.
- The proof bridges geometry, probability, and combinatorics, confirming a combinatorial analog important for discrete mathematics and potentially impacting fields like data science, machine learning, and optimization.
- Although initial attempts involved using ChatGPT to assist, the final and more general proof was provided by Tudose, with the team ultimately not relying on the AI-generated work.