Exploitation of Multiple Replenishing Resources with Uncertainty

TL;DR

This paper studies an optimization problem where a bat maximizes nectar intake from multiple cacti under uncertainty and competition, providing strategies that approximate the optimal with small core sets of cacti.

Contribution

It introduces a purely-stochastic strategy framework and proves bounds on core size needed for near-optimal performance, highlighting the impact of competition intensity.

Findings

A small core of cacti suffices for near-optimal strategies.

The core size bound is asymptotically tight.

More intense competition allows smaller core strategies.

Abstract

We consider an optimization problem in which a (single) bat aims to exploit the nectar in a set of $n$ cacti with the objective of maximizing the expected total amount of nectar it drinks. Each cactus $i \in [n]$ is characterized by a parameter $r_{i} > 0$ that determines the rate in which nectar accumulates in $i$. In every round, the bat can visit one cactus and drink all the nectar accumulated there since its previous visit. Furthermore, competition with other bats, that may also visit some cacti and drink their nectar, is modeled by means of a stochastic process in which cactus $i$ is emptied in each round (independently) with probability $0 < s_i < 1$. Our attention is restricted to purely-stochastic strategies that are characterized by a probability vector $(p_1, \ldots, p_n)$ determining the probability $p_i$ that the bat visits cactus $i$ in each round. We prove that for every $\epsilon > 0$, there exists a purely-stochastic strategy that approximates the optimal purely-stochastic strategy to within a multiplicative factor of $1 + \epsilon$, while exploiting only a small core of cacti. Specifically, we show that it suffices to include at most $\frac{2 (1 - \sigma)}{\epsilon \cdot \sigma}$ cacti in the core, where $\sigma = \min_{i \in [n]} s_{i}$. We also show that this upper bound on core size is asymptotically optimal as a core of a significantly smaller size cannot provide a $(1 + \epsilon)$-approximation of the optimal purely-stochastic strategy. This means that when the competition is more intense (i.e., $\sigma$ is larger), a strategy based on exploiting smaller cores will be favorable.