The price of unfairness in linear bandits with biased feedback. (arXiv:2203.09784v1 [math.ST])

Artificial intelligence is increasingly used in a wide range of decision
making scenarios with higher and higher stakes. At the same time, recent work
has highlighted that these algorithms can be dangerously biased, and that their
results often need to be corrected to avoid leading to unfair decisions. In
this paper, we study the problem of sequential decision making with biased
linear bandit feedback. At each round, a player selects an action described by
a covariate and by a sensitive …

arxiv feedback math price