Simple Mechanisms for Utility Maximization: Approximating Welfare in the I.I.D. Unit-Demand Setting

We investigate the objective of utility maximization from the perspective of Bayesian mechanism design, initiating this direction, and focus on the unit-demand setting where values are i.i.d. across both items and buyers. We take the approach of developing simple, approximately optimal mechanisms, targeting the simplest benchmark of optimal welfare. We give a $(1-1/e)$-approximation when there are more items than buyers, and an O(log(n/m))-approximation when there are more buyers than items, which is tight up to constant factors. We also characterize complexities in this setting that defy our intuition from the welfare and revenue literature, and motivate why coming up with a better benchmark than welfare is a hard problem itself.