Dictionary · A Survey on Bias and Fairness in Machine Learning
L2 — definitions grouped by regulatory framework.
Nouns
5 senses- In-Processing
Techniques that try to modify and change state-of-the-art learning algorithms to remove discrimination during the model training process.
- Equality of Odds
The probability of a person in the positive class being correctly assigned a positive outcome and the probability of a person in a negative class being incorrectly assigned a positive outcome should both be the same for the protected and unprotected group members. In other words, the protected and unprotected groups should have equal rates for true positives and false positives.
- Equality of Opportunity
The probability of a person in positive class being assigned to a positive outcome should be equal for both protected and unprotected group members. In other words, the protected and unprotected groups should have equal true positive rates.
- Individual Fairness
Give similar predictions to similar individuals
- Post-Processing
Performed after training by accessing a holdout set that was not involved during the training of the model. If the algorithm can only treat the learned model as a black box without any ability to modify the training data or learning algorithm, then only post-processing can be used in which the labels assigned by the black-box model initially get reassigned based on a function during the post-processing phase.