2 papers
cs.LG2026
Sampling-Free Privacy Accounting for Matrix Mechanisms under Random Allocation
Jan Schuchardt, Nikita Kalinin
We study privacy amplification for differentially private model training with matrix factorization under random allocation (also known as the balls-in-bins model). Recent work by C…
math.GM2026
A -adic () depth- supercongruence for Gaussian -th power sums over a square
Nikita Kalinin, Faith Shadow Zottor
For an odd prime , define . We study the -adic valuation of these Gaussian power sums for $n\leq p, n=r(p-1…