Neyman-Pearson test for zero-rate multiterminal hypothesis testing

The problem of zero-rate multiterminal hypothesis testing is revisited. A Neyman-Pearson-like test is proposed and its non-asymptotic performance is clarified; for short blocklength, it is numerically examined that the proposed test is superior to a previously known Hoeffding-like test proposed by Han-Kobayashi. For the large deviation regime, it is shown that our proposed test achieves the optimal trade-off between the type I and type II exponents shown by Han-Kobayashi. The information geometry method plays an important role in the analysis as well as the construction of the test.