"Highly sought-after fundamental particles, such as the Higgs boson, only exist for a fraction of a second after being created in high-energy collisions (e.g. ~10-22 s for the Higgs boson). Their presence needs to be inferred from tracking information and energy deposits in large electronic detectors. Often the anticipated detector signature is very rare and could be mimicked by more common background processes. To identify particle decays and increase the efficiency of analyses, ANNs were trained to pick out specific patterns in the large volumes of detector data being generated at a high rate." (emphasis mine)

It concerns me reading stuff like this (one can find similar for the original LIGO detection of gravitational waves) without accompanying qualification. B/c I want to hear them justify why it shouldn't sound like 'we created something that was trained to beg the question ad nauseam'. Obvs on a social trust basis I have every reason to believe these seminal discoveries are precisely as reported. But I'd just like to see what the stats look like - even if I'm probably incapable of really understanding them - that are able to guarantee the validity of an observation when the observation is by definition new, and therefore has never been detected before, and therefore cannot have produced an a priori test set (outside of simulation) baseline to compare against.