-
作者：Wang, Zhaodong; Ouyang, Yanfeng; She, Ruifeng
作者单位：Netflix, Inc.; University of Illinois System; University of Illinois Urbana-Champaign
摘要：This paper presents methods to obtain analytical solutions to a class of continuous traffic equilibrium problems, where continuously distributed customers from a bounded two-dimensional service region seek service from one of several discretely located facilities via the least congested travel path. We show that under certain conditions, the traffic flux at equilibrium, which is governed by a set of partial differential equations, can be decomposed with respect to each facility and solved anal...
-
作者：Davoodi, Mehdi; Katehakis, Michael N.; Yang, Jian
作者单位：Rutgers University System; Rutgers University New Brunswick; Rutgers University Newark; Rutgers University System; Rutgers University New Brunswick; Rutgers University Newark
摘要：We study a dynamic inventory control problem involving fixed setup costs and random demand distributions. With an infinite planning horizon, model primitives including costs and distributions are set to be stationary. Under a given demand distribution, an (s, S) policy has been known to minimize the long-run per-period average cost. Out of the need to model situations involving new products or unencountered economic conditions, however, we depart from the traditional model by allowing the stat...
-
作者：Banerjee, Tathagata; Feinstein, Zachary
作者单位：Washington University (WUSTL); Stevens Institute of Technology
摘要：In this paper, we present formulas for the valuation of debt and equity of firms in a financial network under comonotonic endowments. We demonstrate that the comonotonic setting provides a lower bound and Jensen???s inequality provides an upper bound to the price of debt under Eisenberg-Noe financial networks with bankruptcy costs. Such financial networks encode the interconnection of firms through debt claims. The proposed pricing formulas consider the realized, endogenous recovery rate on de...
-
作者：DeValve, Levi; Pekec, Sasa
作者单位：University of Chicago; Duke University
摘要：We consider ad-supported media platforms with incomplete information about the disutility consumers??? experience from exposure to advertising. We characterize the platform???s optimal menu of subscription prices and advertising quantities in monopoly and competitive settings, revealing insights on key factors influencing market outcomes. In particular, we show that incomplete information on advertising disutility decreases the optimal subscription price for consumers with low advertising disu...
-
作者：Pang, John; Lin, Weixuan; Fu, Hu; Kleeman, Jack; Bitar, Eilyan; Wierman, Adam
作者单位：California Institute of Technology; Cornell University; University of British Columbia; University of Cambridge
摘要：In this paper, we analyze the worst-case efficiency loss of online platform designs under a networked Coumot competition model. Inspired by some of the largest platforms in operation today, we study a variety of platform designs to examine the impacts of market transparency and control on the worst-case efficiency loss of Nash equilibria in networked Coumot games. Our results show that open access designs incentivize increased production toward perfectly competitive levels and limit efficiency...
-
作者：Pearce, Michael Arthur Leopold; Poloczek, Matthias; Branke, Juergen
作者单位：University of Warwick; Amazon.com; University of Warwick
摘要：Bayesian optimization is a powerful tool for expensive stochastic black-box optimization problems, such as simulation-based optimization or machine learning hyperparameter tuning. Many stochastic objective functions implicitly require a random number seed as input. By explicitly reusing a seed, a user can exploit common random numbers, comparing two or more inputs under the same randomly generated scenario, such as a common customer stream in a job shop problem or the same random partition of ...
-
作者：Farahani, Mehdi H.; Dawande, Milind; Janakiraman, Ganesh
作者单位：University of Miami; University of Texas System; University of Texas Dallas
摘要：We study the problem of managing queues in online food-ordering services where customers, who place orders online and pick up at the store, are offered a common quote time, that is, the promised pick-up time minus the time the order is placed. The objective is to minimize the long-run average expected earliness and tardiness cost incurred by the customers. We introduce the family of static threshold policies for managing such queues. A static threshold policy is one that starts serving the fir...
-
作者：Farias, Vivek F.; Gutin, Eli
作者单位：Massachusetts Institute of Technology (MIT); Uber Technologies, Inc.
摘要：Recent years have seen a resurgence of interest in Bayesian algorithms for the multiarmed bandit (MAB) problem, such as Thompson sampling. These algorithms seek to exploit prior information on arm biases. The empirically observed performance of these algorithms makes them a compelling alternative to their frequentist counterparts. Nonetheless, there appears to be a wide range in empirical performance among such Bayesian algorithms. These algorithms also vary substantially in their design (as o...
-
作者：Bensoussan, Alain; Hoe, SingRu (Celine); Kim, Joohyun; Yan, Zhongfeng
作者单位：University of Texas System; University of Texas Dallas; City University of Hong Kong; Texas A&M University System; Jinan University
摘要：The objective of this paper is to study the optimal consumption and portfoliochoice problem of risk-controlled investors who strive to maximize total expected discountedutility of both consumption and terminal wealth. Risk is measured by the variance of terminalwealth, which introduces a nonlinear function of the expected value into the control problem.The control problem presented is no longer a standard stochastic control problem but rather,ameanfield-type control problem. The optimal portfo...
-
作者：Shah, Devavrat; Xie, Qiaomin; Xu, Zhi
作者单位：Massachusetts Institute of Technology (MIT); University of Wisconsin System; University of Wisconsin Madison
摘要：In this work, we consider the popular tree-based search strategy within the framework of reinforcement learning, the Monte Carlo tree search (MCTS), in the context of the infinite-horizon discounted cost Markov decision process (MDP). Although MCTS is believed to provide an approximate value function for a given state with enough simulations, the claimed proof of this property is incomplete. This is because the variant of MCTS, the upper confidence bound for trees (UCT), analyzed in prior work...
