Searcharxiv⌕ Search

arXiv · 2611.21856

Optimal Network Design under Node Failures

Abstract

Networks are frequently constructed before it is determined which nodes will persist. We address the design problem in scenarios where nodes fail independently, links are costly and must be established in advance, and the benefits to surviving nodes diminish with distance. Examples include supplier, interbank, and infrastructure networks. Among all possible networks, the complete, star, and empty configurations are each welfare-optimal within specific ranges of link costs. However, with four or more nodes, failures introduce a range of costs in which every optimal network contains more links than the star but fewer than the complete network. At the lower end of this range, we show that the optimal networks are exactly those obtained by removing a matching from the complete network. We develop an exact algorithm for complete multipartite networks. It finds an optimal network by dynamic programming in a number of operations that grows with the fourth power of the number of nodes, although the class itself grows superpolynomially. For networks with up to six nodes, restricting the design to this class results in no loss under explicit conditions. We also characterize pairwise-stable networks within this class and establish their existence for every level of link cost. Finally, we quantify the cost of misjudging failure risk. Overestimating node survival leads to overbuilding and can yield lower welfare than building nothing, whereas underestimating it never does. Under uncertain failure risk, design should therefore favor conservative reliability estimates.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Michele Lombardi, Francesco Umutcan, Enjun Liu. 2026-11-16. Optimal Network Design under Node Failures. https://searcharxiv.com/abs/2611.21856

Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.

KEEP EXPLORING

Related papers

Simple Communication

We study simple communication in a multidimensional cheap talk model with aligned preferences. An expert communicates with a decision-maker using a score, i.e., an aggregator of a multidimensional state into a one-dimensional message. Even though the expert and the decision-maker share the same payoffs, equilibrium payoffs are lower than those achievable under commitment to a score. Under quadratic-loss utility, any equilibrium score must be linear in the state or a coarsening of a linear score. Finally, for normally distributed states, we characterize the equilibrium linear scores and show they include the ex-ante best and worst linear scores.

econ.TH↗

The Economics of AI Foundation Models: Openness, Competition, and Governance

Foundation model developers choose how open to make their models. Greater openness lowers downstream adaptation costs but can also make competing models easier to use. In a two-period model, an incumbent chooses its license fee and openness, a deployer adapts the model for users, and an entrant offers a competing model. Usage builds model-specific experience that lowers the deployer's future adaptation cost, creating a data flywheel that favors continued use of the incumbent. We identify conditions under which optimal openness falls and then rises as learning strengthens. The incumbent moves from full openness and a high fee to restricting openness for retention, then to a low fee that builds enough usage and experience to retain the deployer despite greater openness. A full-openness mandate can induce the incumbent to abandon this low-fee strategy. The higher fee can reduce first-period adaptation despite greater openness. An openness trap arises when this welfare loss exceeds the second-period gain from easier adaptation of the entrant's model. The trap can disappear when experience transfers across models. Full integration eliminates the fee between the incumbent and deployer but commits deployment to the incumbent, potentially forgoing a rival that is cheaper to adapt. Integration raises their joint profit, consumer surplus, and welfare when learning is sufficiently strong, but can lower them when learning is weak. A small usage subsidy can benefit both developers, the deployer, and consumers. It can instead induce a fee increase larger than the subsidy and reduced openness, raising the deployer's net payment per use and adaptation costs. The resulting reduction in adaptation lowers welfare even before subsidy expenditure is deducted. These findings connect AI policy to developers' pricing and openness responses and the portability of deployment experience.

econ.TH↗

Convex Order and Moment Persuasion

We study Bayesian persuasion with multidimensional states when the sender's payoff depends on the distribution of posterior means. The choice of information policy in such an environment reduces to an optimization over distributions of posterior means dominated by the prior in convex order. We study extreme points of this feasible set using results on irreducible-component structure from martingale transport theory. When the prior assigns zero mass to the union of the relative boundaries of the non-singleton components, then on each such component the distribution of posterior means is dominated by the prior in convex order. In such cases, extreme points must satisfy affine independence of the conditional posterior-mean support in almost every component. Without boundary negligibility, local affine independence remains necessary but is not sufficient. Unlike in the one-dimensional case, in environments with multidimensional states, extreme points can have uncountably many lower-dimensional components that are not singletons.

econ.TH↗