1期

最新文章

• The Role of Lookahead and Approximate Policy Evaluation in Reinforcement Learning with Linear Value Function Approximation AI翻译
  作者：Winnicki, Anna; Lubars, Joseph; Livesay, Michael; Srikant, R.
  作者单位：University of Illinois System; University of Illinois Urbana-Champaign; University of Illinois System; University of Illinois Urbana-Champaign; United States Department of Energy (DOE); Sandia National Laboratories; University of Illinois System; University of Illinois Urbana-Champaign
  摘要：Function approximation is widely used in reinforcement learning to handle the computational difficulties associated with very large state spaces. However, function approximation introduces errors that may lead to instabilities when using approximate dynamic programming techniques to obtain the optimal policy. Therefore, techniques such as lookahead for policy improvement and m-step rollout for policy evaluation are used in practice to improve the performance of approximate dynamic programming ...
• Achieving Efficiency in Black-Box Simulation of Distribution Tails with Self-Structuring Importance Samplers AI翻译
  作者：Deo, Anand; Murthy, Karthyek
  作者单位：Indian Institute of Management (IIM System); Indian Institute of Management Bangalore; Singapore University of Technology & Design
  摘要：This paper presents a novel importance sampling (IS) scheme for estimating distribution tails of performance measures modeled with a rich set of tools, such as linear programs, integer linear programs, piecewise linear/quadratic objectives, feature maps specified with deep neural networks, etc. The conventional approach of explicitly identifying efficient changes of measure suffers from feasibility and scalability concerns beyond highly stylized models because of their need to be tailored intr...
• Improving the Security of United States Elections with Robust Optimization AI翻译
  作者：Crimmins, Braden L.; Halderman, J. Alex; Sturt, Bradley
  作者单位：University of Michigan System; University of Michigan; University of Illinois System; University of Illinois Chicago; University of Illinois Chicago Hospital
  摘要：For more than a century, election officials across the United States have inspected voting machines before elections using a procedure called logic and accuracy testing (LAT). This procedure consists of election officials casting a test deck of ballots into each voting machine and confirming the machine produces the expected vote total for each candidate. We bring a scientific perspective to LAT by introducing the first formal approach to designing test decks with rigorous security guarantees....
• A Stationary Infinite-Horizon Supply Contract Under Asymmetric Inventory Information AI翻译
  作者：Bensoussan, Alain; Sethi, Suresh; Wang, Shouqiang
  作者单位：University of Texas System; University of Texas Dallas; City University of Hong Kong
  摘要：We consider a decentralized supply chain in which a supplier sells goods to a retailer facing general random demand over an infinite horizon. The retailer satisfies the demand to the extent of the inventory on hand. The retailer has private information about the retailer's stock in each period, and the supplier offers the retailer a supply contract menu to account for the information asymmetry. We obtain a necessary condition for optimizing a long-term stationary truth-telling contract under g...
• Gaussian Process-Based Random Search for Continuous Optimization via Simulation AI翻译
  作者：Wang, Xiuxian; Hong, L. Jeff; Jiang, Zhibin; Shen, Haihui
  作者单位：Shanghai Jiao Tong University; Fudan University; Fudan University; Shanghai Jiao Tong University
  摘要：Random search is an important category of algorithms to solve continuous optimization via simulation problems. To design an efficient random search algorithm, the handling of the triple E (i.e., exploration, exploitation and estimation) is critical. The first two E's refer to the design of sampling distribution to balance explorative and exploitative searches, whereas the third E refers to the estimation of objective function values based on noisy simulation observations. In this paper, we pro...
• The When and How of Delegated Search AI翻译
  作者：Zorc, Sasa; Tsetlin, Ilia; Hasija, Sameer; Chick, Stephen E.
  作者单位：University of Virginia; INSEAD Business School; INSEAD Business School
  摘要：Firms often outsource search processes, such as the acquisition of real estate, new technologies, or talent. To ensure the efficacy of such delegated search, firms need to carefully design incentive contracts to attenuate the ill effects of agency issues. We model this problem using a dynamic principal-agent framework, embedding the standard sequential search model. The optimal contract pays the agent a fixed per-period fee plus a bonus for finding a suitable alternative. The bonus size is def...
• Conic Mixed-Binary Sets: Convex Hull Characterizations and Applications AI翻译
  作者：Kilinc-Karzan, Fatma; Kucukyavuz, Simge; Lee, Dabeen; Shafieezadeh-Abadeh, Soroosh
  作者单位：Carnegie Mellon University; Northwestern University; Korea Advanced Institute of Science & Technology (KAIST)
  摘要：We consider a general conic mixed-binary set where each homogeneous conic constraint j involves an affine function of independent continuous variables and an epigraph variable associated with a nonnegative function, fj, of common binary variables. Sets of this form naturally arise as substructures in a number of applications, including mean-risk optimization, chance-constrained problems, portfolio optimization, lot sizing and scheduling, fractional programming, variants of the best subset sele...
• Behavior-Aware Queueing: The Finite-Buffer Setting with Many Strategic Servers AI翻译
  作者：Zhong, Yueyang; Gopalakrishnan, Ragavendran; Ward, Amy R.
  作者单位：University of Chicago; Queens University - Canada
  摘要：Service system design is often informed by queueing theory. Traditional queueing theory assumes that servers work at constant speeds. That is reasonable in computer science and manufacturing contexts. However, servers in service systems are people, and in contrast to machines, the incentives created by design decisions influence their work speeds. We study how server work speed is affected by managerial decisions concerning (i) how many servers to staff and how much to pay them and (ii) whethe...
• Application-Driven Learning: A Closed-Loop Prediction and Optimization Approach Applied to Dynamic Reserves and Demand Forecasting AI翻译
  作者：Dias, Joaquim; Street, Alexandre; Homem-de-Mello, Tito; Munoz, Francisco D.
  作者单位：Universidad Adolfo Ibanez
  摘要：Decision making is generally modeled as sequential forecast-decision steps with no feedback, following an open-loop approach. For instance, in the electricity sector, system operators use the forecast-decision approach followed by ad hoc rules to determine reserve requirements and biased net load forecasts to guard the system against renewable generation and demand uncertainty. Such procedures lack technical formalism to minimize operating and reliability costs. We present a new closed-loop fr...
• Optimal Routing Under Demand Surges: The Value of Future Arrival Rates AI翻译
  作者：Chen, Jinsheng; Dong, Jing; Shi, Pengyi
  作者单位：Agency for Science Technology & Research (A*STAR); A*STAR - Singapore Institute of Manufacturing Technology (SIMTech); Columbia University; Purdue University System; Purdue University
  摘要：Motivated by the growing availability of advanced demand forecast tools, we study how to use future demand information in designing routing strategies in queueing sys-tems under demand surges. We consider a parallel server system operating in a nonstation-ary environment with general time-varying arrival rates. Servers are cross-trained to help nonprimary customer classes during demand surges. However, such flexibility comes with various operational costs, such as a loss of efficiency and inco...