The sample complexity of binary hypothesis testing scales as 1/JSD for the optimal LLR classifier and as 1/JSD² for a majority-vote classifier, giving JSD an operational interpretation in terms of data requirements.
J.et al.Generative Adversarial Nets.Advances in Neural Information Processing Systems27,2672–2680 (2014)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IT 1years
2026 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
Sample complexity bounds for the Jensen-Shannon divergence
The sample complexity of binary hypothesis testing scales as 1/JSD for the optimal LLR classifier and as 1/JSD² for a majority-vote classifier, giving JSD an operational interpretation in terms of data requirements.