3期

最新文章

• Optimal and Differentially Private Data Acquisition: Central and Local Mechanisms AI翻译
  作者：Fallah, Alireza; Makhdoumi, Ali; Malekian, Azarakhsh; Ozdaglar, Asuman
  作者单位：Massachusetts Institute of Technology (MIT); Duke University; University of Toronto
  摘要：We consider a platform's problem of collecting data from privacy sensitive users to estimate an underlying parameter of interest. We formulate this question as a Bayesianoptimal mechanism design problem, in which an individual can share their (verifiable) data in exchange for a monetary reward or services, but at the same time has a (private) heterogeneous privacy cost which we quantify using differential privacy. We consider two popular differential privacy settings for providing privacy guar...
• Robust Dynamic Assortment Optimization in the Presence of Outlier Customers AI翻译
  作者：Chen, Xi; Krishnamurthy, Akshay; Wang, Yining
  作者单位：New York University; University of Texas System; University of Texas Dallas
  摘要：We consider the dynamic assortment optimization problem under the multinomial logit model with unknown utility parameters. The main question investigated in this paper is model mis-specification under the e-contamination model, which is a fundamental model in robust statistics and machine learning. In particular, throughout a selling horizon of length T, we assume that customers make purchases according to a well-specified underlying multinomial logit choice model in a (1 - e)-fraction of the ...
• An Improved Analysis of LP-Based Control for Revenue Management AI翻译
  作者：Chen, Guanting; Li, Xiaocheng; Ye, Yinyu
  作者单位：University of North Carolina; University of North Carolina Chapel Hill; University of North Carolina School of Medicine; Imperial College London; Stanford University
  摘要：In this paper, we study a class of revenue-management problems, where the decision maker aims to maximize the total revenue subject to budget constraints on multiple types of resources over a finite horizon. At each time, a new order/customer/bid is revealed with a request of some resource(s) and a reward, and the decision maker needs to either accept or reject the order. Upon the acceptance of the order, the resource request must be satisfied, and the associated revenue (reward) can be collec...
• Wasserstein Distributionally Robust Optimization and Variation Regularization AI翻译
  作者：Gao, Rui; Chen, Xi; Kleywegtc, Anton J.
  作者单位：University of Texas System; University of Texas Austin; New York University; University System of Georgia; Georgia Institute of Technology
  摘要：Wasserstein distributionally robust optimization (DRO) is an approach to optimization under uncertainty in which the decision maker hedges against a set of probability distributions, specified by a Wasserstein ball, for the uncertain parameters. This approach facilitates robust machine learning, resulting in models that sustain good performance when the data are to some extent different from the training data. This robustness is related to the well-studied effect of regularization. The connect...
• Dynamic Placement in Refugee Resettlement AI翻译
  作者：Ahani, Narges; Golz, Paul; Procaccia, Ariel D.; Teytelboym, Alexander; Trapp, Andrew C.
  作者单位：Bank of America Corporation; Harvard University; University of Oxford; Worcester Polytechnic Institute; Worcester Polytechnic Institute
  摘要：Employment outcomes of resettled refugees depend strongly on where they are initially placed in the host country. Each week, a resettlement agency is allocated a set of refugees by the U.S. government. The agency must place these refugees in its local affiliates while respecting the affiliates' annual capacities. We develop an allocation system that recommends where to place an incoming refugee family to improve total employment success. Our algorithm is based on two-stage stochastic programmi...
• Proximal Reinforcement Learning: Efficient Off-Policy Evaluation in Partially Observed Markov Decision Processes AI翻译
  作者：Bennett, Andrew; Kallus, Nathan
  作者单位：Cornell University
  摘要：In applications of offline reinforcement learning to observational data, such as in healthcare or education, a general concern is that observed actions might be affected by unobserved factors, inducing confounding and biasing estimates derived under the assumption of a perfect Markov decision process (MDP) model. Here we tackle this by considering off-policy evaluation in a partially observed MDP (POMDP). Specifically, we consider estimating the value of a given target policy in an unknown POM...
• Risk Sensitivity and Firm Power: Price Competition with Mean2-Variance Profit Objective Under Multinomial Logit Demand AI翻译
  作者：Li, Hongmin; Webster, Scott
  作者单位：Arizona State University; Arizona State University-Tempe
  摘要：This paper is the first in the literature to address a risk-sensitive price competition under the multinomial logit choice model, with each participating firm maximizing a riskadjusted profit objective. We find that, at equilibrium, a subset of firms earns a positive profit, whereas others are driven to zero profit, contrasting with the risk-neutral equilibrium in which all firms earn a positive profit regardless of quality and cost. We identify a power index-the ratio of effective product att...
• Optimal Dynamic Control of an Epidemic AI翻译
  作者：Kruse, Thomas; Strack, Philipp
  作者单位：University of Wuppertal; Yale University
  摘要：We analyze how to optimally engage in social distancing in order to minimize the spread of an infectious disease. We identify conditions under which any optimal policy is single peaked (i.e., first engages in increasingly more social distancing and subsequently decreases its intensity). We show that an optimal policy might substantially delay measures that decrease the transmission rate to create herd immunity and that engaging in social distancing suboptimally early can increase the number of...
• Data-Driven Optimization with Distributionally Robust Second Order Stochastic Dominance Constraints AI翻译
  作者：Peng, Chun; Delage, Erick
  作者单位：Beijing Jiaotong University; Universite de Montreal; HEC Montreal; Universite de Montreal; HEC Montreal
  摘要：Optimization with stochastic dominance constraints has recently received an increasing amount of attention in the quantitative risk management literature. Instead of requiring that the probabilistic description of the uncertain parameters be exactly known, this paper presents a comprehensive study of a data-driven formulation of the distributionally robust second order stochastic dominance constrained problem (DRSSDCP) that hinges on using a type-1 Wasserstein ambiguity set. This formulation a...
• Endogenous Credit, Business Cycle, and Portfolio Selection AI翻译
  作者：Choi, Kyoung Jin; Koo, Hyeng Keun; Lim, Byung Hwa; Yoo, Jane
  作者单位：University of Calgary; Ajou University; Pohang University of Science & Technology (POSTECH); Sungkyunkwan University (SKKU)
  摘要：We study a continuous-time model of consumption and portfolio selection of an agent with a limited ability to commit to a debt contract in which the credit limit is endogenously determined. We consider the case where the agent borrows against future income and/or collateral assets. We also study the determination of the credit limit in a general equilibrium model. We derive the credit limit in closed form. The credit limit is smaller than the natural limit because of limited commitment and an ...