arXiv subjects

Yehua Wei

Publications and source records attributed to Yehua Wei.

10 recordsLinked to original sources

arXiv2026

Longest-Queue Policies in Dynamic Matching with Abandonments

We study a dynamic two-way matching market in which agents of heterogeneous types arrive over time, wait to be matched according to a compatibility network, and abandon the system if their random patience expires. The decision maker collects match rewards and pays holding and abandonment costs, and seeks to maximize the expected net value over a finite horizon. We evaluate policies by their approximation ratio against a fluid benchmark, the optimal value of a linear-programming relaxation that upper bounds the expected net value of every policy. Under a general position gap (GPG) condition, which requires the unique optimal dual solution of this linear program to remain optimal under small perturbations of the arrival rates, we propose a longest-queue policy. The policy first uses this dual solution to identify a set of good match types and a set of critical agent types. Upon each arrival, it considers only good matches and matches the arrival to the critical partner with the longest queue subject to availabilities. In the high-volume regime, in which arrival rates are scaled by s, we prove that the policy attains an approximation ratio of at least 1-O(1/s) at every scale s >= 1 and every horizon T >= 1, so the ratio tends to 1 as s goes to infinity.

math.OC↗

arXiv2026

Optimality of Random Regular Graphs in Sparse Network Designs

The problems of designing sparse networks arise frequently in resource allocation and operations research. In production systems, for example, sparse process flexibility designs are used to handle uncertain demand effectively: the goal is to construct the sparsest bipartite graph between supply and demand that still achieves an expected fulfilled demand comparable to that of a fully flexible system. In middle-mile transportation, sparse delivery-route subgraphs that sustain large matchings after random node deletions help reduce delivery costs; here, the goal is to design the sparsest graph whose maximum matching size remains comparable to that of the fully connected graph under node deletions. The design of sparse networks has been studied extensively, with state-of-the-art results providing order-wise optimal designs for both bipartite and unipartite networks (Chen et al., 2015; Feng et al., 2024). However, identifying designs that achieve the sharp theoretical limit -- where the average degree asymptotically matches the lower bound of any graph to achieve a given loss level, has remained open. In this paper, we prove that the random regular graph achieves this sharp optimal condition in both bipartite and unipartite settings. Numerical experiments further validate this optimality. Our results highlight a practical guideline for sparse flexibility networks: designs that combine degree regularity with dispersed edge placement can achieve optimal performance under uncertainty.

cs.DS↗

arXiv2026

Online Pandora's Box for Contextual LLM Cascading

Motivated by Large Language Model (LLM) cascading, we propose an online contextual Pandora's Box model for adaptively querying and selecting LLM APIs. In each period, a decision-maker observes a request context and faces a two-phase decision problem. In the query phase, the decision-maker sequentially queries APIs, where each query reveals a generated output and the decision-maker incurs an (output-dependent) cost. In the selection phase, the decision-maker selects one of the generated outputs to deploy and observes only the downstream reward of the deployed output. This output-mediated feedback structure differs from classical online contextual Pandora's Box models, in which opening a box directly reveals its reward. Rather than estimating the full conditional output and cost distributions of each API, we directly model the reservation index and develop a learning approach for the query phase. Specifically, we impose a parametric structure on the contextual reservation index functions induced by the classical Weitzman's policy. Our policy combines generalized method of moments (GMM) type estimation of these reservation indices with UCB-style confidence bounds for both these indices and the shared output-level reward evaluator. Under regularity conditions, we prove that the resulting policy achieves dimension-dependent $\widetilde O(\sqrt T)$ cumulative regret over a horizon of $T$ periods.

cs.AI↗

arXiv2026

COOPA: A Modular LLM Agent Architecture for Operations Research Problems

Operations Research (OR) provides a rigorous framework for high-stakes decision-making, but effective OR modeling requires substantial domain knowledge, mathematical abstraction, and solver expertise. Recent LLM-based systems automate parts of this pipeline, yet remain limited by low accuracy on complex problems, opaque outputs, and narrow solver support. We propose COOPA (COoperative OPerations Agent), a modular LLM-agent architecture for interpretable and scalable OR decision support. It combines three components: iterative confidence-based modeling, which generates multiple candidate formulations, self-evaluates them across modeling dimensions, and selects one using a max-min confidence criterion; element-level provenance and confidence explanations, which link variables, parameters, constraints, and objectives to quoted source text and provide an audit trail for human verification; and multi-solver routing to specialized optimizer agents for different OR problem classes. Across three OR benchmarks, eight LLM backbones, and four baselines under identical conditions, COOPA achieves the best macro-average accuracy on six of eight backbones and improves over the strongest baseline by up to 6.7 percentage points. A within-system ablation isolates the contribution of iterative confidence-based modeling, while additional analyses and case studies illustrate the value of source traceability and multi-solver dispatch.

cs.LG↗

arXiv2022

Latency Guarantee for Ubiquitous Intelligence in 6G: A Network Calculus Approach

With the gradual deployment of 5G and the continuous popularization of edge intelligence (EI), the explosive growth of data on the edge of the network has promoted the rapid development of 6G and ubiquitous intelligence (UbiI). This article aims to explore a new method for modeling latency guarantees for UbiI in 6G given 6G's extremely stochastic nature in terahertz (THz) environments, THz channel tail behavior, and delay distribution tail characteristics generated by the UBiI random component, and to find the optimal solution that minimizes the end-to-end (E2E) delay of UbiI. In this article, the arrival curve and service curve of network calculus can well characterize the stochastic nature of wireless channels, the tail behavior of wireless systems and the E2E service curve of network calculus can model the tail characteristic of the delay distribution in UbiI. Specifically, we first propose demands and challenges facing 6G, edge computing (EC), edge deep learning (DL), and UbiI. Then, we propose the hierarchical architecture, the network model, and the service delay model of the UbiI system based on network calculus. In addition, two case studies demonstrate the usefulness and effectiveness of the network calculus approach in analyzing and modeling the latency guarantee for UbiI in 6G. Finally, future open research issues regarding the latency guarantee for UbiI in 6G are outlined.

cs.NI↗

arXiv2021

Reinforcement Learning for Flexibility Design Problems

Flexibility design problems are a class of problems that appear in strategic decision-making across industries, where the objective is to design a ($e.g.$, manufacturing) network that affords flexibility and adaptivity. The underlying combinatorial nature and stochastic objectives make flexibility design problems challenging for standard optimization methods. In this paper, we develop a reinforcement learning (RL) framework for flexibility design problems. Specifically, we carefully design mechanisms with noisy exploration and variance reduction to ensure empirical success and show the unique advantage of RL in terms of fast-adaptation. Empirical results show that the RL-based method consistently finds better solutions compared to classical heuristics.

cs.LG↗

arXiv2017

Dynamic Load-Balancing Vertical Control for Large-Scale Software-Defined Internet of Things

As the global Internet of things increasingly is popular with consumers and business environment, network flow management has become an important topic to optimize the performance on Internet of Things. The rigid existing Internet of things (IoT) architecture blocks current traffic management technology to provide a real differentiated service for large-scale IoT. Software-defined Internet of Things (SD-IoT) is a new computing paradigm that separates control plane and data plane, and enables centralized logic control. In this paper, we first present a general framework for SD-IoT, which consists of two main components: SD-IoT controllers and SD-IoT switches. The controllers of SD-IoT uses resource pooling technology, and the pool is responsible for the centralized control of the entire network. The switches of SD-IoT integrate with the gateway functions, which is responsible for data access and forwarding. The SD-IoT controller pool is designed as a vertical control architecture, which includes the main control layer and the base control layer. The controller (main controller) of the main control layer interacts upward with the application layer, interacts with the base control layer downwards, and the controller (base controller) of the basic control layer interacts with the data forwarding layer. We propose a dynamic balancing algorithm of the main controller based on election mechanism and a dynamic load balancing algorithm of the basic controller based on the balanced delay, respectively. The experimental results show that the dynamic balancing algorithm based on the election mechanism can ensure the consistency of the messages between the main controllers, and the dynamic load balancing algorithm based on the balanced delay can balance between these different workloads in the basic controllers.

cs.NI↗

arXiv2016

An adaptive $O(\log n)$-optimal policy for the online selection of a monotone subsequence from a random sample

Given a sequence of $n$ independent random variables with common continuous distribution, we propose a simple adaptive online policy that selects a monotone increasing subsequence. We show that the expected number of monotone increasing selections made by such a policy is within $O(\log n)$ of optimal. Our construction provides a direct and natural way for proving the $O(\log n)$-optimality gap. An earlier proof of the same result made crucial use of a key inequality of Bruss and Delbaen (2001) and of de-Poissonization.

math.PR↗

arXiv2012

Belief Propagation for Min-cost Network Flow: Convergence and Correctness

Message passing type algorithms such as the so-called Belief Propagation algorithm have recently gained a lot of attention in the statistics, signal processing and machine learning communities as attractive algorithms for solving a variety of optimization and inference problems. As a decentralized, easy to implement and empirically successful algorithm, BP deserves attention from the theoretical standpoint, and here not much is known at the present stage. In order to fill this gap we consider the performance of the BP algorithm in the context of the capacitated minimum-cost network flow problem - the classical problem in the operations research field. We prove that BP converges to the optimal solution in the pseudo-polynomial time, provided that the optimal solution of the underlying problem is unique and the problem input is integral. Moreover, we present a simple modification of the BP algorithm which gives a fully polynomial-time randomized approximation scheme (FPRAS) for the same problem, which no longer requires the uniqueness of the optimal solution. This is the first instance where BP is proved to have fully-polynomial running time. Our results thus provide a theoretical justification for the viability of BP as an attractive method to solve an important class of optimization problems.

cs.DM↗

arXiv2007

A criterion for the half-plane property

We establish a convenient necessary and sufficient condition for a multiaffine real polynomial to be stable, and use it to verify that the half-plane property holds for seven small matroids that resisted the efforts of Choe, Oxley, Sokal, and Wagner [5].

math.CO↗