-
作者：Zhou, Jiaqi; Ryzhov, Ilya O.
作者单位：University System of Maryland; University of Maryland College Park; University System of Maryland; University of Maryland College Park
摘要：We derive a new optimal sampling budget allocation for belief models based on linear regression with continuous covariates, where the expected response is interpreted as the value of the covariate vector, and an error occurs if a lower-valued vector is falsely identified as being better than a higher-valued one. Our allocation optimizes the rate at which the probability of error converges to zero using a large deviations theoretic characterization. This is the first large deviations-based opti...
-
作者：Kim, Taeho; Kim, Kyoung-Kuk; Song, Eunhye
作者单位：Hong Kong University of Science & Technology; Korea Advanced Institute of Science & Technology (KAIST); University System of Georgia; Georgia Institute of Technology
摘要：We consider an expected-value ranking and selection (R&S) problem where all k solutions' simulation outputs depend on a common parameter whose uncertainty can be modeled by a distribution. We define the most probable best (MPB) to be the solution that has the largest probability of being optimal with respect to the distribution and design an efficient sequential sampling algorithm to learn the MPB when the parameter has a finite support. We derive the large deviations rate of the probability o...
-
作者：Fan, Lin; Glynn, Peter W.
作者单位：Northwestern University; Stanford University
摘要：Much of the literature on optimal design of bandit algorithms is based on minimization of expected regret. It is well known that algorithms that are optimal over certain exponential families can achieve expected regret that grows logarithmically in the number of trials at a rate specified by the Lai-Robbins lower bound. In this paper, we show that when one uses such optimized algorithms, the resulting regret distribution necessarily has a very heavy tail, specifically that of a truncated Cauch...
-
作者：Ebert, Sebastian; Karehnke, Paul
作者单位：Ruprecht Karls University Heidelberg; heSam Universite; ESCP Business School
摘要：Prudence is widely known for inducing saving in response to future risk- precautionary saving. In this paper, we revisit this important implication by providing definitions of first- and second-order prudence that parallel Segal and Spivak's [Segal U, Spivak A (1990) First order versus second order risk aversion. J. Econom. Theory 51:111-125] definitions of first- and second-order risk aversion. Second-order prudence means that the decision-maker is almost prudent-neutral for small risks, wher...
-
作者：Dong, Jing; Ibrahim, Rouba
作者单位：Columbia University; University of London; University College London
摘要：Size-based scheduling has been extensively studied yet almost exclusively in single-server queues with infinitely patient jobs and perfectly known service times. Much less is known about its performance in many-server queues, particularly under noisy service-time information. In this paper, we derive theoretical results that quantify the performance of the nonpreemptive shortest-job-first (SJF) policy in many-server queues with abandonment and noisy service-time estimates. In particular, we co...
-
作者：Chen, Li; Fu, Chenyi; Si, Fan; Sim, Melvyn; Xiong, Peng
作者单位：University of Sydney; Northwestern Polytechnical University; National University of Singapore
摘要：Robust optimization presents a compelling methodology for optimization under uncertainty, providing a practical, ambiguity-averse evaluation of risk when the probability distribution is encapsulated by an ambiguity set. We introduce the moment-dispersion ambiguity set, an improvement on the moment-based set, enabling separate characterization of a random variable's location, dispersion, and support. To describe dispersion, we define the dispersion characteristic function, capturing complex att...
-
作者：Vardi, Shai; Haskell, William
作者单位：Purdue University System; Purdue University
摘要：Scarce resources are often shared between different stakeholders and need to be scheduled fairly among them. In this paper, we study the utility loss resulting from requiring the schedule to be fair, using the price of fairness-the ratio of the utility attainable with and without fairness restrictions. We focus on envy-freeness, an intuitive and well-studied fairness notion. We derive tight bounds on the price of fairness as a function of the problem parameters: the number of agents, the time ...
-
作者：Zhong, Yueyang; Birge, John R.; Ward, Amy R.
作者单位：University of Chicago
摘要：The multiclass many-server queue with abandonment (specifically, the multiclass GI/GI/N+GI queue) is a canonical model for service systems. One key operational question is how to schedule; that is, how to choose the customer that a newly available server will serve. The scheduling question is of fundamental importance because scheduling determines which customer classes have more abandonments. However, even though there is much work on scheduling in queueing systems, there is comparatively les...
-
作者：Gallego, Guillermo; Li, Anran
作者单位：The Chinese University of Hong Kong, Shenzhen; Chinese University of Hong Kong
摘要：In this paper, we operationalize the random consideration set (RCS) choice model proposed by Manzini and Mariotti which assumes that consumers make purchase decisions based on a fixed preference ordering and random consideration sets drawn from independent attention probabilities. We provide a necessary condition, a sufficient condition, and a fast algorithm to estimate preference ordering and attention probabilities from sales transaction data, thereby uniquely identifying the model parameter...
-
作者：Long, Zhenghua; Zhang, Hailun; Zhang, Jiheng; Zhang, Zhe George
作者单位：Nanjing University; Shenzhen Research Institute of Big Data; The Chinese University of Hong Kong, Shenzhen; Hong Kong University of Science & Technology; Western Washington University; Simon Fraser University
摘要：We study the optimal control of a queueing model with a single customer class and heterogeneous server pools. The main objective is to strike a balance between the holding cost of the queue and the operating costs of the server pools. We introduce a target-allocation policy, which assigns higher priority to the queue or pools without enough customers for general cost functions. Although we can prove its asymptotic optimality, implementation requires solving a nonlinear optimization problem. Wh...
