Zhiwei Steven Wu

Zhiwei Steven Wu

Auto-generated profile, based on publicly available information

Zhiwei Steven Wu

Carnegie Mellon University

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 March 2022 in Operations Research
Verified
Journal Article
Closed access
YMYishay Mansour
Yishay Mansour
ASAleksandrs Slivkins
Aleksandrs Slivkins
VSVasilis Syrgkanis
Vasilis Syrgkanis
ZWZhiwei Steven Wu
Zhiwei Steven Wu
In a wide range of recommendation systems, self-interested individuals (“agents”) make decisions over time, using information revealed by other agents in the past, and producing information that may help agents in the future. Each agent would like to exploit the best action given the current information but would prefer the previous agents to explore various alternatives to collect information. A social planner, by means of a well-designed recommendation policy, can incentivize the agents to balance exploration and exploitation in order to maximize social welfare or some other objective. The recommendation policy can be modeled as a multiarmed bandit algorithm under Bayesian incentivecompatibility (BIC) constraints. This line of work has received considerable attention in the “economics and computation” community. Although in prior work, the planner interacts with a single agent at a time, the present paper allows the agents to affect one another directly in a shared environment. The agents now face two sources of uncertainty: what is the environment, and what would the other agents do? We focus on “explorable” actions: those that can be recommended by some BIC policy. We show how the principal can identify and explore all such actions.

12 June 2023
Verified
Conference Paper
Open access
LGLuke Guerdan
Luke Guerdan
ACAmanda Coston
Amanda Coston
ZWZhiwei Steven Wu
Zhiwei Steven Wu
KHKenneth Holstein
Kenneth Holstein
A growing literature on human-AI decision-making investigates strategies for combining human judgment with statistical models to improve decision-making. Research in this area often evaluates proposed improvements to models, interfaces, or workflows by demonstrating improved predictive performance on “ground truth’’ labels. However, this practice overlooks a key difference between human judgments and model predictions. Whereas humans commonly reason about broader phenomena of interest in a decision – including latent constructs that are not directly observable, such as disease status, the “toxicity” of online comments, or future “job performance” – predictive models target proxy labels that are readily available in existing datasets. Predictive models’ reliance on simplistic proxies for these nuanced phenomena makes them vulnerable to various sources of statistical bias. In this paper, we identify five sources of target variable bias that can impact the validity of proxy labels in human-AI decision-making tasks. We develop a causal framework to disentangle the relationship between each bias and clarify which are of concern in specific human-AI decision-making tasks. We demonstrate how our framework can be used to articulate implicit assumptions made in prior modeling work, and we recommend evaluation strategies for verifying whether these assumptions hold in practice. We then leverage our framework to re-examine the designs of prior human subjects experiments that investigate human-AI decision-making, finding that only a small fraction of studies examine factors related to target variable bias. We conclude by discussing opportunities to better address target variable bias in future research.

7 February 2025 in ACM Transactions on Economics and Computation
Verified
Journal Article
Open access
GAGuy Aridor
Guy Aridor
YMYishay Mansour
Yishay Mansour
ASAleksandrs Slivkins
Aleksandrs Slivkins
SWSteven Wu
Steven Wu
Most online platforms learn from interactions with users and engage in exploration : making potentially suboptimal choices to acquire new information. We study the interplay between exploration and competition : how such platforms balance the exploration for learning and competition for users. We consider a stylized duopoly in which two firms face the same multi-armed bandit problem. Users arrive one by one and choose between the two firms, so that each firm makes progress on its bandit problem only if it is chosen. We study whether competition incentivizes the adoption of better algorithms. We find that stark competition disincentivizes exploration, leading to low welfare. However, weaker competition incentivizes better exploration algorithms and increases welfare. We investigate two channels for weakening the competition: stochastic user choice models and a first-mover advantage. Our findings speak to the competition–innovation relationship and the first-mover advantage in the digital economy.

12 June 2023
Verified
Conference Paper
Open access
LGLuke Guerdan
Luke Guerdan
ACAmanda Coston
Amanda Coston
KHKenneth Holstein
Kenneth Holstein
ZWZhiwei Steven Wu
Zhiwei Steven Wu
Across domains such as medicine, employment, and criminal justice, predictive models often target labels that imperfectly reflect the outcomes of interest to experts and policymakers. For example, clinical risk assessments deployed to inform physician decision-making often predict measures of healthcare utilization (e.g., costs, hospitalization) as a proxy for patient medical need. These proxies can be subject to outcome measurement error when they systematically differ from the target outcome they are intended to measure. However, prior modeling efforts to characterize and mitigate outcome measurement error overlook the fact that the decision being informed by a model often serves as a risk-mitigating intervention that impacts the target outcome of interest and its recorded proxy. Thus, in these settings, addressing measurement error requires counterfactual modeling of treatment effects on outcomes. In this work, we study intersectional threats to model reliability introduced by outcome measurement error, treatment effects, and selection bias from historical decision-making policies. We develop an unbiased risk minimization method which, given knowledge of proxy measurement error properties, corrects for the combined effects of these challenges. We also develop a method for estimating treatment-dependent measurement error parameters when these are unknown in advance. We demonstrate the utility of our approach theoretically and via experiments on real-world data from randomized controlled trials conducted in healthcare and employment domains. As importantly, we demonstrate that models correcting for outcome measurement error or treatment effects alone suffer from considerable reliability limitations. Our work underscores the importance of considering intersectional threats to model validity during the design and evaluation of predictive models for decision support.

29 February 2020 in ACM Transactions on Economics and Computation
Verified
Journal Article
Open access
ARAaron Roth
Aaron Roth
ASAleksandrs Slivkins
Aleksandrs Slivkins
JUJonathan Ullman
Jonathan Ullman
ZWZhiwei Steven Wu
Zhiwei Steven Wu
We study the problem of a seller dynamically pricing d distinct types of indivisible goods, when faced with the online arrival of unit-demand buyers drawn independently from an unknown distribution. The goods are not in limited supply, but can only be produced at a limited rate and are costly to produce. The seller observes only the bundle of goods purchased at each day, but nothing else about the buyer’s valuation function. Our main result is a dynamic pricing algorithm for optimizing welfare (including the seller’s cost of production) that runs in time and a number of rounds that are polynomial in d and the approximation parameter. We are able to do this despite the fact that (i) the price-response function is not continuous, and even its fractional relaxation is a non-concave function of the prices, and (ii) the welfare is not observable to the seller. We derive this result as an application of a general technique for optimizing welfare over divisible goods, which is of independent interest. When buyers have strongly concave, Hölder continuous valuation functions over d divisible goods, we give a general polynomial time dynamic pricing technique. We are able to apply this technique to the setting of unit-demand buyers despite the fact that in that setting the goods are not divisible, and the natural fractional relaxation of a unit-demand valuation is not strongly concave. To apply our general technique, we introduce a novel price randomization procedure that has the effect of implicitly inducing buyers to “regularize” their valuations with a strongly concave function. Finally, we also extend our results to a limited-supply setting in which the supply of each good cannot be replenished.

11 June 2024 in ACM SIGMETRICS Performance Evaluation Review
Verified
Journal Article
Open access
KHKeegan Harris
Keegan Harris
AAAnish Agarwal
Anish Agarwal
CPChara Podimata
Chara Podimata
ZWZhiwei Steven Wu
Zhiwei Steven Wu
We consider the problem of decision-making using panel data, in which a decision-maker gets noisy, repeated measurements of multiple units (or agents). We consider the setup used in synthetic control methods, where there is a pre-intervention period when the principal observes the outcomes of each unit, after which the principal uses these observations to assign a treatment to each unit. Unlike this classical setting, we permit the units generating the panel data to be strategic, i.e. units may modify their pre-intervention outcomes in order to receive a more desirable intervention. The principal's goal is to design a strategyproof intervention policy, i.e. a policy that assigns units their utility-maximizing interventions despite their potential strategizing. We first identify a necessary and sufficient condition under which a strategyproof intervention policy exists, and provide a strategyproof mechanism with a simple closed form when one does exist. Along the way, we prove impossibility results for strategic multiclass classification, a natural extension of the well-studied (binary) strategic classification problem to the multiclass setting, which may be of independent interest. When there are two interventions, we establish that there always exists a strategyproof mechanism, and provide an algorithm for learning such a mechanism. For three or more interventions, we provide an algorithm for learning a strategyproof mechanism if there exists a sufficiently large gap in the principal's rewards between different interventions. Finally, we empirically evaluate our model using real-world panel data collected from product sales over 18 months. We find that our methods compare favorably to baselines which do not take strategic interactions into consideration, even in the presence of model misspecification.