Latest Seminars

'Now or Later?': When to Deploy Qualification Screening in Open-Bid Auction for Re-Sourcing
Dr Qi Chen, George, London Business School

This paper considers a re-sourcing setting in which a qualified supplier (the incumbent) and multiple suppliers which have not yet been qualified (the entrants) compete in an open-bid descending auction for a single-supplier contract. Due to the risk of supplier nonperformance, the buyer only awards the contract to a qualified supplier; meanwhile, the buyer can conduct supplier qualification screening at a cost, to verify whether the entrant suppliers can perform the contract. Conventionally, the buyer would screen entrants before running an auction, i.e., the pre-qualification strategy (PRE). We explore an alternative approach called post-qualification strategy (POST), in which the buyer first runs an auction and then conducts qualification screenings based on the suppliers' auction bids. Our characterization of the dynamic structure of the suppliers' equilibrium bidding strategy enables the calculation of the buyer's expected cost under POST, which is computationally intractable without this characterization. We derive analytical conditions under which POST is cheaper than PRE, and also use a comprehensive numerical study to quantify the benefit of POST. We find that using the cheaper option between PRE and POST not only provides significant cost-savings over the conventional PRE-only approach but also captures the majority of the benefit an optimal mechanism can offer over PRE. Our results highlight the practical benefit of POST.

Date 13.08.2021
Time 4:00 - 5:15 PM
Venue Zoom ID: 941 1769 1148 (passcode 607141)

Incentive-Compatible Assortment Optimization
Dr Antoine Désir, INSEAD

Online marketplaces, such as Amazon, Alibaba, or Google Shopping, allow sellers to promote their products by charging them for the right to be displayed on top of organic search results. In this paper, we study the problem of designing auctions for promoted products and highlight some new challenges emerging from the interplay of two unique features: substitution effects and information asymmetry. The presence of substitution effects, which we capture by assuming that consumers choose sellers according to a multinomial logit model, implies that the probability a seller is chosen depends on the assortment of sellers displayed alongside. Additionally, sellers may hold private information about how their own products match consumers’ interests, which the platform can elicit to make better assortment decisions. We first show that the first-best allocation, i.e., the welfare-maximizing assortment in the absence of private information, cannot be implemented truthfully in general. Thus motivated, we initiate the study of incentive-compatible assortment optimization by characterizing prior-free and prior-dependent mechanisms, and quantifying the worst-case social cost of implementing truthful assortment mechanisms. An important finding is that the worst-case social cost of implementing truthful mechanisms can be high when the number of sellers is large. Structurally, we show that optimal mechanisms may need to downward distort the efficient allocation both at the top and the bottom. This is joint work with Santiago Balseiro.

Date 06.08.2021
Time 4:00 - 5:15 PM
Venue Zoom ID: 999 0394 5595 (passcode 882030)

Scaling Up Battery Swapping Services in Cities: A Joint Location and Repairable-Inventory Model
Dr Wei Qi, McGill University

Battery swapping for electric vehicle refueling is reviving and thriving. Despite a captivating sustainable future where swapping batteries will be as convenient as refueling gas today, a tension is mounting in practice (beyond the traditional “range anxiety” issue): On one hand, it is desirable to maximize battery proximity and availability to customers. On the other hand, power grids for charging depleted batteries are not accessible everywhere. To reconcile this tension, some cities are embracing an emerging infrastructure network: Decentralized swapping stations replenish charged batteries from centralized charging stations. It remains unclear how to design such a network, or whether transitioning into this paradigm will save batteries which are environmentally detrimental. In this paper, we model this new urban infrastructure network. This task is complicated by non-Poisson swaps (observed from real data), and by the intertwined stochastic operations of swapping, charging, stocking and circulating batteries among swapping and charging stations. We show that these complexities can be captured by analytical models. We next propose a new location-inventory model for citywide deployment of hub charging stations, which jointly determines the location, allocation and reorder quantity decisions with a non-convex non- concave objective function. We solve this problem exactly and efficiently by exploiting the hidden submodularity and combining constraint-generation and parameter-search techniques. Even for solving convexified problems, our algorithm brings a speedup of at least three orders of magnitude relative to Gurobi solver. The major insight is twofold: Centralizing battery charging may harm cost-efficiency and battery asset-lightness; however, this finding is reversed if foreseeing that decentralized charging will have limited access to grids permitting fast charging. We also identify planning and operational flexibilities brought by centralized charging. In a broader sense, this work deepens our understanding about how mobility and energy are coupled in future smart cities.

Date 23.07.2021
Time 9:00 – 10:15 am
Venue Zoom ID: 943 8935 2374 (passcode 227609)

Dynamic Batch Learning in High-Dimensional Sparse Linear Contextual Bandits
Dr Zhengyuan Zhou, New York University

We study the problem of dynamic batch learning in high-dimensional sparse linear contextual bandits, where a decision maker can only adapt decisions at a batch level. In particular, the decision maker, only observing rewards at the end of each batch, dynamically decides how many individuals to include in the next batch (at the current batch's end) and what personalized action-selection scheme to adopt within the batch. Such batch constraints are ubiquitous in a variety of practical contexts, including personalized product offerings in marketing and medical treatment selection in clinical trials. We characterize the fundamental learning limit in this problem via a novel lower bound analysis and provide a simple, exploration-free algorithm that uses the LASSO estimator, which achieves the minimax optimal performance characterized by the lower bound (up to log factors). To our best knowledge, our work provides the first inroad into a rigorous understanding of dynamic batch learning with high-dimensional covariates.

Date 16.07.2021
Time 9:00 – 10:05 am
Venue Zoom ID: 924 9031 1025 (passcode 813461)

Algorithmic Processes of Social Alertness and Social Transmission: How Bots Disseminate Information on Twitter
Prof. Elena KARAHANNA, University of Georgia

Date 08.07.2021
Time 9:00 am - 10:30 am (Hong Kong Time)
Venue Zoom

Sharing and Sourcing of Online Misinformation
Prof. Susan BROWN, The University of Arizona

Date 02.07.2021
Time 9:00 am - 10:30 am (Hong Kong Time)
Venue Zoom

COVID-19 Impacts on Work and Life
Prof. Viswanath VENKATESH, Virginia Tech

Date 30.06.2021
Time 9:00 am - 10:30 am (Hong Kong Time)
Venue Zoom

Choice Overload with Search Cost and Anticipated Regret: Field Evidence and Theoretical Framework
Dr Jiankun Sun, Imperial College London

We examine the impact of assortment size on consumer choice behavior with both empirical evidence and theoretical explanation. We first conduct a large-scale field experiment in online retail to causally examine how consumers' click and purchase behavior changes as the number of products in a choice set increases. There, we document a non-monotonic relationship between the assortment size and consumer choice. We then develop a two-stage choice model that incorporates consumers’ search cost and anticipated regret to explain our findings in the field experiment. We also conduct numerical experiments to investigate the implications of our model for companies' optimal assortment decisions. Our results suggest that our two-stage choice model leads to smaller optimal assortments containing products of higher expected utilities and lower prices on average than the classical multinomial logit (MNL) choice model.

Date 25.06.2021
Time 4:00 - 5:15 pm
Venue Zoom ID: 986 5486 1916 (passcode 930247)

Delaying Informed Consent: An Empirical Investigation of Mobile Apps’ Upgrade Decisions
Prof. Raveesh MAYYA, Assistant Professor, New York University

Date 23.06.2021
Time 9:00 am - 10:30 am (Hong Kong Time)
Venue Zoom

Multi-Item Online Order Fulfillment in a Two-Layer Network
Dr Linwei Xin, University of Chicago

The boom of e-commerce in the globe in recent years has expedited the expansion of fulfillment infrastructures by e-retailers. While e-retailers are building more and more mini-warehouses close to end customers to offer faster delivery service than ever, the associated fulfillment costs have skyrocketed. In this paper, we study a real-time fulfillment problem in a two-layer RDC-FDC distribution network that has been implemented in practice by major e-retailers. In such a network, the upper layer contains larger regional distribution centers (RDCs) and the lower layer contains smaller front distribution centers (FDCs). We allow order split: an order can be split and fulfilled from multiple warehouses at an additional cost. The objective is to minimize the routine fulfillment costs. We study real-time algorithms with performance guarantees in both settings with and without demand forecasts. We also complement our theoretical results by conducting a numerical study by using real data from Alibaba.

This is joint work with Xinshang Wang (Alibaba) and Yanyang Zhao (Chicago Booth).

Date 11.06.2021
Time 09:30 - 10:45 am
Venue Zoom ID: 982 9972 1714 (passcode 315647)

Robust Active Learning for Personalization
Dr Chaithanya Bandi, National University of Singapore

We consider the problem faced by an e-retailer that needs to display a limited set of products to a customer with no prior information. In this context, the e-retailer is allowed to query preferences in order to inform its display. A standard approach to this problem follows a two-step approach: First, estimate the preferences of the customer using a choice model, and then optimize the product display. While this approach is applicable to many settings with stationary customer preferences, this is not applicable to scenarios with changing customer preferences. In this paper, we develop a novel product-driven online framework for efficiently learning customer preferences using a structured questionnaire design. We demonstrate that our approach provably outperforms state-of-the-art methods which focus on eliciting the preference vector. Further, we formulate a robust algorithm for eliciting the optimal display set when the customer responses are noisy. 

We establish theoretical foundations for our question-design mechanism and develop efficiency guarantees for our product-driven algorithm. We also present results of our implementation on a real data set obtained from a major fashion retailer. We demonstrate that we are able to efficiently and customer preferences to inform the optimal product display, and outperform existing approaches based on "estimate, then optimize".

Joint work with Yam Huo (NU), and based on work with Jonathan Amar and Nikos Trichakis (MIT).

Date 14.05.2021
Time 10:30 - 11:45 am
Venue Zoom ID: 947 8336 6439 (passcode 911538)

Joint Statistics Seminar - High Dimensional Forecast Combinations Under Latent Structures
Professor Zhentao SHI, The Chinese University of Hong Kong

This paper presents a novel high-dimensional forecast combination estimator in the presence of many forecasts and potential latent group structures.  The new algorithm, which we call ℓ2-relaxation, minimizes the squared ℓ2-norm of the weight vector subject to a relaxed version of the first-order conditions, instead of minimizing the mean squared forecast error as those standard optimal forecast combination procedures.  A proper choice of the tuning parameter achieves bias and variance trade-off, and incorporates as special cases the simple average (equal-weight) strategy and the conventional optimal weighting scheme. When the variance-covariance (VC) matrix of the individual forecast errors exhibits latent group structures -- a block equicorrelation matrix plus a VC for idiosyncratic noises, ℓ2-relaxation delivers combined forecasts with roughly equal within-group weights.  Asymptotic optimality of the new method is established by exploiting the duality between the sup-norm restriction and the high-dimensional sparse ℓ1-norm penalization.  Excellent finite sample performance of our method is demonstrated in Monte Carlo simulations.  Its wide applicability is highlighted in three real data examples concerning empirical applications of microeconomics, macroeconomics, and finance.

Based on joint work with Liangjun Su, Tian Xie.

Date 14.05.2021
Time 9:00 – 10:00 am
Venue Zoom ID 922 7350 7265

Wage Elasticity of Labor Supply in Real-Time Ridesharing Markets: An Empirical Analysis
Prof. Liangfei Qiu, PricewaterhouseCoopers Associate Professor, University of Florida

Date 12.05.2021
Time 9:00 am - 10:30 am (Hong Kong Time)
Venue Zoom

Optimal Policies and Heuristics To Match Supply With Demand For Online Retailing
Dr Yun-Fong Lim, Singapore Management University

We consider an online retailer selling multiple products to multiple zones over a single period. The retailer orders the products from a single supplier and stores them at multiple warehouses. At the start of the selling period, the retailer determines the order quantities of the products and their storage quantities at each warehouse subject to its capacity constraint. At the end of the period, after knowing the demands, the retailer determines the retrieval quantities from each warehouse to fulfill the demands. The retailer's objective is to maximize her expected profit. For the single- zone case, we solve the problem optimally. The optimal retrieval policy is a greedy policy. We design a polynomial-time algorithm to determine the optimal storage policy, which preserves a nested property: Among all non-empty warehouses, a smaller-index warehouse contains all the products stored in a larger-index warehouse. The optimal ordering policy is a newsvendor-type policy. The problem becomes intractable analytically if there are multiple zones and we propose an efficient heuristic to solve it. This heuristic involves a non-trivial hybrid approximation of the second- stage expected profit. Numerical experiments using both synthetic data and real data from a major fashion online retailer in Asia suggest that our heuristic outperforms state-of-the-art approaches with significantly less computational time. With flexible fulfillment, our heuristic improves the efficiency by 28% on average compared to a dedicated policy adopted by theretailer.

Date 07.05.2021
Time 10:30 - 11:45 am
Venue Zoom ID: 927 9987 9236 (passcode 632818)

Probabilistic Sequential Shrinking: A Best Arm Identification Algorithm for Stochastic Bandits with Corruptions
Dr Wang-Chi Cheung, National University of Singapore

Motivated by the issues of fraudulent clicks in online recommendation systems and contaminated samples in medical trials, we consider a best arm identification (BAI) problem for stochastic bandits with adversarial corruptions. The goal is to identify the best arm with a fixed number of pulls (which are also known as time steps), in the presence of an adversary who can corrupt the stochastic outcomes of the arms.

We design a novel randomized algorithm, PROBABILISTIC SEQUENTIAL SHRINKING (PSS), which is agnostic to the amount of corruptions. In the absence of corruptions, our proposed algorithm achieves the state-of- the-art performance guarantee. In the presence of corruptions, we construct settings where the state-of- the-art BAI algorithm (Karnin et al. 2013) fails to identify the best arm with probability at least 0.5, whereas PSS identifies the best arm with high probability. En route, we demonstrate the importance of randomized sampling for mitigating the impact of corruptions.

In addition, we identify the amount of corruptions per step (CPS) to be a crucial parameter that characterizes the possibility of BAI. When the CPS is below a certain threshold, PSS identifies the best arm with high probability. Otherwise, the optimality gap of the identified arm degrades gracefully with the CPS, while there is no guarantee on the probability of identifying the best arm. We demonstrate the necessity of such a bifurcation, by showing that BAI is impossible when the CPS is above a certain threshold.

Date 30.04.2021
Time 10:30 - 11:45 am
Venue Zoom ID: 961 8647 4544 (passcode 405397)