For the measurement of $N_s$ signals in $N$ events rigorous confidence bounds of the signal probability $p$ were established in a classical paper by Clopper and Pearson [Biometrica 26 (1934) 404]. Here, I generalize their bounds to the situation where a neural network (or similar device) tags signal data with likelihood $p_s$ and background data with likelihood $p_b$.

Back to Program

Back to ACAT Web page