E-variables and tests of randomness for distribution classes
Georgii Potapov, Yuri Kalnishkan · 2026
Read E-variables and tests of randomness for distribution classes by Georgii Potapov, Yuri Kalnishkan (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-2603-02492.txt.
Description
Georgii Potapov and Yuri Kalnishkan build hypothesis tests from e-variables in this 2026 paper. E-variables are a newer tool for testing claims with strong error control. They grew from algorithmic randomness tests, a concept from information theory. Building good e-variables for whole families of distributions is the practical challenge.
Their approximability method constructs e-variables for popular distribution classes. The idea approximates a complex class by simpler pieces where e-variables are known. That yields explicit tests usable in applications. The same construction gives an explicit randomness test for a distribution class. Such tests judge whether data look truly random under the class. The work links modern safe testing back to its algorithmic roots.
Catalog record
| CREATOR | Georgii Potapov, Yuri Kalnishkan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 53057 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.02492