Log Loss
A metric for evaluating the quality of probabilistic predictions that imposes harsher penalties on extreme incorrect predictions; the closer a probability is to 0 or 1 when an error occurs, the greater the loss.
Timeline (last 90 days)
The core mechanism of Calibration Court is to write each AI prediction into a hash-chain ledger before the outcome is revealed, and to score it using the Brier score, log loss, and calibration curves.
The paper establishes for the first time a regret lower bound for online learning of quantum states under the logarithmic loss.
All Facts (2)
The core mechanism of Calibration Court is to write each AI prediction into a hash-chain ledger before the outcome is revealed, and to score it using the Brier score, log loss, and calibration curves.
50%UnverifiedThe paper establishes for the first time a regret lower bound for online learning of quantum states under the logarithmic loss.
50%