Online Fair Allocations with Binary Valuations and Beyond

Abstract: In an online fair allocation problem, a sequence of indivisible items arrives online and needs to be allocated to offline agents immediately and irrevocably. In our paper, we study the online allocation of either goods or chores. We adapt popular fairness notions, including envy-freeness up to one item (EF1) and maximin share fairness (MMS), and use utilitarian social welfare (USW) to measure efficiency. For both settings of items, we present a series of positive results regarding the existence of fair and efficient allocations with widely studied classes of binary and bivalued valuation/cost functions. Additionally, we complement our results by constructing some counterexamples to establish our results as among the best guarantees possible.