-
作者：Kannan, Rohit; Bayraksan, Guezin; Luedtke, James R.
作者单位：Virginia Polytechnic Institute & State University; University System of Ohio; Ohio State University; University of Wisconsin System; University of Wisconsin Madison; University of Wisconsin System; University of Wisconsin Madison
摘要：We study optimization for data-driven decision making when we have observations of the uncertain parameters within an optimization model together with concurrent observations of covariates. The goal is to choose a decision that minimizes the expected cost conditioned on a new covariate observation. We investigate two data-driven frameworks that integrate a machine learning prediction model within a stochastic programming sample average approximation (SAA) for approximating the solution to this...
-
作者：Ba, Wenjia; Lin, Tianyi; Zhang, Jiawei; Zhou, Zhengyuan
作者单位：University of British Columbia; Columbia University; New York University
摘要：We consider online no-regret learning in unknown games with bandit feedback, where each player can only observe its reward at each time-determined by all players' current joint action-rather than its gradient. We focus on the class of smooth and strongly monotone games and study optimal no-regret learning therein. Leveraging self-concordant barrier functions, we first construct a new bandit learning algorithm and show that it root ffiffiffi achieves the single-agent optimal regret of Theta ( n...
-
作者：Cai, Yang; Oikonomou, Argyris
作者单位：Yale University
摘要：We study the problem of selling n heterogeneous items to a single buyer, whose values for different items are dependent. Under arbitrary dependence, others show that no simple mechanism can achieve a nonnegligible fraction of the optimal revenue even with only two items. We consider the setting where the buyer's type is drawn from a correlated distribution that can be captured by a Markov random field (MRF), one of the most prominent frameworks for modeling high-dimensional distributions with ...
-
作者：Cai, Biao; Zhang, Jingfei; Sun, Will Wei
作者单位：City University of Hong Kong; Emory University; Purdue University System; Purdue University
摘要：We consider the problem of jointly modeling and clustering populations of tensors by introducing a high-dimensional tensor mixture model with heterogeneous covariances. To effectively tackle the high dimensionality of tensor objects, we employ plausible dimension reduction assumptions that exploit the intrinsic structures of tensors, such as low rankness in the mean and separability in the covariance. In estimation, we develop an efficient high-dimensional expectation conditional maximization ...
-
作者：Pesenti, Silvana M.; Jaimungal, Sebastian; Saporito, Yuri F.; Targino, Rodrigo S.
作者单位：University of Toronto; University of Oxford; Getulio Vargas Foundation
摘要：We define and develop an approach for risk budgeting allocation-a risk diversification portfolio strategy-where risk is measured using a dynamic time-consistent risk measure. For this, we introduce a notion of dynamic risk contributions that generalize the classical Euler contributions, which allows us to obtain dynamic risk contributions in a recursive manner. We prove that for the class of coherent dynamic distortion risk measures, the risk allocation problem may be recast as a sequence of s...
-
作者：Chen, Boxiao; Shi, Cong
作者单位：University of Illinois System; University of Illinois Chicago; University of Illinois Chicago Hospital; University of Miami
摘要：We consider a periodic-review dual-sourcing inventory system in which the expedited supplier is faster and more costly, whereas the regular supplier is slower and cheaper. Under full demand distributional information, it is well known that the optimal policy is extremely complex but the celebrated Tailored Base-Surge (TBS) policy performs near optimally. Under such a policy, a constant order is placed at the regular source in each period, whereas the order placed at the expedited source follow...
-
作者：Li, Weiyuan; Rusmevichientong, Paat; Topaloglu, Huseyin
作者单位：University of Southern California
摘要：When modeling the demand in revenue management systems, a natural approach is to focus on a canonical interval of time, such as a week, so that we forecast the demand over each week in the selling horizon. Ideally, we would like to use random variables with general distributions to model the demand over each week. The current demand can give a signal for the future demand, so we also would like to capture the dependence between the demands over different weeks. Prevalent demand models in the l...
-
作者：den Hertog, Dick; Pauphilet, Jean; Pham, Yannick; Sainte-Rose, Bruno; Song, Baizhi
作者单位：University of Amsterdam; University of London; London Business School
摘要：Increasing ocean plastic pollution is irreversibly harming ecosystems and human economic activities. We partner with a nonprofit organization and use optimization to help clean up oceans from plastic faster. Specifically, we optimize the route of their plastic collection system in the ocean to maximize the quantity of plastic collected over time. We formulate the problem as a longest path problem in a well-structured graph. However, because collection directly impacts future plastic density, t...
-
作者：Braouezec, Yann; Kiani, Keyvan
作者单位：IESEG School of Management; Centre National de la Recherche Scientifique (CNRS); CNRS - Institute for Humanities & Social Sciences (INSHS); Universite de Lille; emlyon business school; Centre National de la Recherche Scientifique (CNRS); Ecole Normale Superieure de Lyon (ENS de LYON); Universite Claude Bernard Lyon 1; Universite Jean Monnet; Universite Lyon 2
摘要：We offer a stress test framework in which interaction between regulated banks occurs through the impact they may have on asset prices when they deleverage. Because banks are constrained to maintain their risk-based capital ratio higher than a threshold, the deleveraging problem yields a generalized game in which the solvency constraint of each bank depends on the decisions of the others. We analyze the game under microprudential but also under macroprudential regulation. Microprudential regula...
-
作者：Yang, Yu
作者单位：State University System of Florida; University of Florida
摘要：In this paper, we propose an innovative variable fixing strategy called deep Lagrangian underestimate fi xing (DeLuxing). It is a highly effective approach for removing unnecessary variables in column-generation (CG)-based exact methods used to solve challenging discrete optimization problems commonly encountered in various industries, including vehicle routing problems (VRPs). DeLuxing employs a novel linear programming (LP) formulation with only a small subset of the enumerated variables, wh...
