The computer models learn by reinforcement, updating when their predictions about outcomes are wrong.
The flaw in this argument lies in the assumption that in every case there must be a single legally correct outcome, with other outcomes being wrong.
None of this is to say that Dennett's preferred outcome is wrong.
But if the outcome is wrong — no matter why — an audit should guarantee a large known chance of correcting the outcome.
Risk-limiting audits (RLAs), developed in 2007, guarantee that if the outcome is wrong, there is a large chance that the audit will correct the record before the results are official.
"Risk-limiting audits" (RLAs), developed in 2007, guarantee that if the outcome is wrong, there is a large chance that the audit will correct the record before the results are official.
Rather than asking what percentage of the ballots you should count by hand to be confident of finding at least one discrepancy if the outcome is wrong, he said, it makes more sense to ask how confident you are that the outcome is right, in light of the discrepancies you find.
Abstract: Risk-limiting audits (RLAs) offer a statistical guarantee: if a full manual tally of the paper ballots would show that the reported election outcome is wrong, an RLA has a known minimum chance of leading to a full manual tally.
"He has already proved what a great manager [at Porto] he is and sometimes in life you do the right things and the outcome is wrong.
Conversely, the use of a surrogate end point instead of a final outcome is wrong by a methodological point of view and can lead to draw misleading conclusions.
