Title:Fair Division in a Variable Setting

Abstract:We study the classic problem of fairly dividing a set of indivisible items among a set of agents and consider the popular fairness notion of envy-freeness up to one item (EF1). While in reality, the set of agents and items may vary, previous works have studied static settings, where no change can occur in the system. We initiate and develop a formal model to understand fair division under the variable input setting: here, there is an EF1 allocation that gets disrupted because of the loss/deletion of an item, or the arrival of a new agent, resulting in a near-EF1 allocation. The objective is to perform a sequence of transfers of items between agents to regain EF1 fairness by traversing only via near-EF1 allocations. We refer to this as the EF1-Restoration problem.
In this work, we present algorithms for the above problem when agents have identical monotone valuations, and items are either all goods or all chores. Both of these algorithms achieve an optimal number of transfers (at most $m/n$, where $m$ and $n$ are the number of items and agents respectively) for identical additive valuations. Next, we consider a valuation class with graphical structure, introduced by Christodoulou et al. (EC'23), where each item is valued by at most two agents, and hence can be seen as an edge between these two agents in a graph. Here, we consider EF1 orientations on (multi)graphs - allocations in which each item is allocated to an agent who values it. While considering EF1 orientations on multi-graphs with additive binary valuations, we present an optimal algorithm for the EF1-Restoration problem. Finally, for monotone binary valuations, we show that the problem of deciding whether EF1-Restoration is possible is PSPACE-complete.