Additively Competitive Secretaries

TL;DR

This paper introduces a regret-based evaluation framework for the secretary problem, showing that certain pricing algorithms can achieve significantly lower regret than the classical worst-case competitive ratio, with tight bounds and broader objectives.

Contribution

It proposes a novel regret minimization framework for secretary algorithms and establishes tight bounds for pricing curve policies within this framework.

Findings

Pricing curves achieve at most 0.25 regret, tight within the class.

Best-only pricing curves achieve at most 0.19 regret, with a lower bound of 0.171.

No policy can guarantee a regret better than 0.152.

Abstract

In the secretary problem, a set of secretary candidates arrive in a uniformly random order and reveal their values one by one. A company, who can only hire one candidate and hopes to maximize the expected value of its hire, needs to make irrevocable online decisions about whether to hire the current candidate. The classical framework of evaluating a policy is to compute its worst-case competitive ratio against the optimal solution in hindsight, and there the best policy -- the ``$1/e$ law'' -- has a competitive ratio of $1/e$. We propose an alternative evaluation framework through the lens of regret -- the worst-case additive difference between the optimal hindsight solution and the expected performance of the policy, assuming that each value is normalized between $0$ and $1$. The $1/e$ law for the classical framework has a regret of $1 - 1/e \approx 0.632$; by contrast, we show that the class of ``pricing curves'' algorithms can guarantee a regret of at most $1/4 = 0.25$ (which is tight within the class), and the class of ``best-only pricing curves'' algorithms can guarantee a regret of at most $0.190$ (with a lower bound of $0.171$). In addition, we show that in general, no policy can give a regret guarantee better than $0.152$. Finally, we discuss other objectives in our regret-minimization framework, such as selecting the top-$k$ candidates for $k > 1$, or maximizing revenue during the selection process.