On a Gradient Approach to Chebyshev Center Problems with Applications to Function Learning

Abhinav Raghuvanshi [email protected], Mayank Baranwal [email protected], Debasish Chatterjee [email protected] · 2026

Read On a Gradient Approach to Chebyshev Center Problems with Applications to Function Learning by Abhinav Raghuvanshi [email protected], Mayank Baranwal [email protected], Debasish Chatterjee [email protected] (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-06434.txt.

Description

Finding the smallest ball that covers a set of functions is a classic problem with modern uses in learning. Abhinav Raghuvanshi and 2 co-authors introduce gradOL in a 2026 paper, the first gradient-based framework for Chebyshev center problems. Reformulating them had resisted gradient methods because the constraints are semi-infinite.

Their reformulation recasts the problem as a finite max-min optimization that gradients can handle. Automatic differentiation supplies precise numerical gradients, keeping the method stable and scalable. Under strong convexity of the ambient norm, gradOL provably recovers optimal centers and computes the radius directly. On 34 benchmark problems it improved accuracy and efficiency. Extended to general convex semi-infinite programming, it beat the leading solver by up to four thousand times on a standard library.

Catalog record

CREATORAbhinav Raghuvanshi [email protected], Mayank Baranwal [email protected], Debasish Chatterjee [email protected]
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE60 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-06434

Browse the full library · About the Tower of Babel