Repeated Prophet Inequality with Near-optimal Bounds

TL;DR

This paper studies a two-phase Prophet Inequality problem where an adversary chooses item sequences, and the goal is to develop algorithms that maximize the expected value relative to an offline optimal, achieving near-optimal competitive ratios.

Contribution

The paper introduces a new two-phase Prophet Inequality model with adversarial sequences and provides algorithms with near-optimal competitive ratios, improving upon basic methods.

Findings

Basic algorithms achieve at most 0.450 ratio.

Proposed algorithm achieves at least 0.495 ratio.

No algorithm can surpass a 0.502 ratio, establishing near-optimality.

Abstract

In modern sample-driven Prophet Inequality, an adversary chooses a sequence of $n$ items with values $v_1, v_2, \ldots, v_n$ to be presented to a decision maker (DM). The process follows in two phases. In the first phase (sampling phase), some items, possibly selected at random, are revealed to the DM, but she can never accept them. In the second phase, the DM is presented with the other items in a random order and online fashion. For each item, she must make an irrevocable decision to either accept the item and stop the process or reject the item forever and proceed to the next item. The goal of the DM is to maximize the expected value as compared to a Prophet (or offline algorithm) that has access to all information. In this setting, the sampling phase has no cost and is not part of the optimization process. However, in many scenarios, the samples are obtained as part of the decision-making process. We model this aspect as a two-phase Prophet Inequality where an adversary chooses a sequence of $2n$ items with values $v_1, v_2, \ldots, v_{2n}$ and the items are randomly ordered. Finally, there are two phases of the Prophet Inequality problem with the first $n$-items and the rest of the items, respectively. We show that some basic algorithms achieve a ratio of at most $0.450$. We present an algorithm that achieves a ratio of at least $0.495$. Finally, we show that for every algorithm the ratio it can achieve is at most $0.502$. Hence our algorithm is near-optimal.