Referee-Meta-Learning for Fast Adaptation of Locational Fairness

arXiv:2402.13379v1 Announce Type: new
Abstract: When dealing with data from distinct locations, machine learning algorithms tend to demonstrate an implicit preference of some locations over the others, which constitutes biases that sabotage the spatial fairness of the algorithm. This unfairness can easily introduce biases in subsequent decision-making given broad adoptions of learning-based solutions in practice. However, locational biases in AI are largely understudied. To mitigate biases over locations, we propose a locational meta-referee (Meta-Ref) to oversee the few-shot meta-training and …

abstract algorithm algorithms arxiv biases cs.cy cs.lg data decision fairness locations machine machine learning machine learning algorithms making meta meta-learning solutions spatial the algorithm type