Selected work

Representative Papers

Evaluating for the long term: Learnings from industry

Aug 08, 2026

Short-term online experiments often fail to accurately predict long-term business outcomes, potentially leading to decisions misaligned with strategic objectives. Drawing on insights from industry expert workshops, this work proposes a design principle for surrogate metrics centered on decision utility rather than solely on unbiasedness, advocating that interpretable, experiment-driven simple surrogates outperform complex black-box models. Through expert consensus synthesis, surrogate metric analysis, and comparative evaluation of experimental versus observational data, the study systematically outlines a methodology for constructing effective surrogates and uncovers a critical relationship between the stability of long-term effects and surrogate validity. While underscoring the irreplaceable value of high-quality long-term experimentation, the research also delineates core challenges and practical guidelines for surrogate learning in real-world settings.

0 citationsRead paper

Non-Exclusive Notifications for Ride-Hailing at Lyft II: Simulations and Marketplace Analysis

Mar 22, 2026

This study addresses the inefficiencies of traditional one-to-one exclusive ride dispatching, which often suffers from matching delays due to driver rejections or response timeouts, thereby reducing platform efficiency. The authors propose a non-exclusive dispatch mechanism that broadcasts each order to multiple drivers and develop an integrated evaluation framework combining a constrained welfare-maximization optimization model, large-scale discrete-event simulations based on real Lyft data, and macroscopic equilibrium analysis. Their findings reveal a fundamental trade-off between matching speed and service quality, demonstrating that non-exclusive dispatch significantly reduces matching time, decreases passenger abandonment, and improves both order completion rates and average service quality. Furthermore, they introduce a conservative notification strategy that mitigates over-allocation of high-value drivers, thereby enhancing long-term market efficiency.

0 citationsRead paper

Non-Exclusive Notifications for Ride-Hailing at Lyft I: Single-Cycle Approximation Algorithms

Mar 22, 2026

This study addresses the notification set selection problem for ride-hailing platforms under uncertainty in driver acceptance, aiming to maximize social welfare under non-exclusive notification mechanisms. For the “First-Accept” (FA) and “Best-Accept” (BA) protocols, the work presents the first PTAS and a constant-factor approximation algorithm, respectively. It establishes the monotone submodularity of the BA objective function and designs a demand oracle that surpasses the classical $1 - 1/e$ approximation barrier. Under homogeneous acceptance probabilities, an exact polynomial-time algorithm is further derived for the BA setting. Rigorous theoretical analysis provides strong approximation guarantees, while experiments on both synthetic data and real-world Lyft-calibrated datasets demonstrate the superior performance of the proposed algorithms in terms of matching quality and response efficiency.

0 citationsRead paper

Informal and Privatized Transit: Incentives, Efficiency and Coordination

Feb 11, 2026

This study addresses the critical yet often overlooked role of informal private transit in urban mobility, whose profit-driven operations are frequently neglected in planning, leading to systemic inefficiencies. The work proposes the first analytically tractable game-theoretic model of fully privatized informal transit, capturing strategic interactions between operators and cost-sensitive passengers. It introduces a tight bound on the Price of Anarchy and designs a Stackelberg routing framework coupled with route-specific cross-subsidization mechanisms. Theoretical analysis reveals that while decentralized decision-making incurs significant but bounded efficiency losses, minimal centralized intervention suffices to substantially mitigate them. Simulations based on real-world data from Nalasopara, India, empirically validate the effectiveness of the proposed mechanisms.

0 citationsRead paper

Post Launch Evaluation of Policies in a High-Dimensional Setting

Dec 30, 2024

In large-scale online platforms with hundreds of millions of users, conventional A/B testing is infeasible for post-hoc policy evaluation under sparse interventions, leading to severe estimation bias. To address this, we propose a two-stage debiased counterfactual estimation framework: (1) a covariate-based nearest-neighbor matching stage to construct high-fidelity control units and mitigate interpolation bias; and (2) a high-dimensional supervised learning stage—using XGBoost or neural networks—to model treatment effects while systematically diagnosing and correcting machine-learning-induced estimation bias. This is the first scalable solution enabling synthetic control methods to operate effectively under ultra-large-scale, sparse-intervention settings. Evaluated across six real-world online experiments, our method significantly improves causal effect estimation accuracy and reduces policy decision error rates by 42%. It has been deployed operationally to support closed-loop decision-making across multiple business units.

0 citationsRead paper

Recent publications

Latest Papers

Evaluating for the long term: Learnings from industry

Aug 08, 2026

Short-term online experiments often fail to accurately predict long-term business outcomes, potentially leading to decisions misaligned with strategic objectives. Drawing on insights from industry expert workshops, this work proposes a design principle for surrogate metrics centered on decision utility rather than solely on unbiasedness, advocating that interpretable, experiment-driven simple surrogates outperform complex black-box models. Through expert consensus synthesis, surrogate metric analysis, and comparative evaluation of experimental versus observational data, the study systematically outlines a methodology for constructing effective surrogates and uncovers a critical relationship between the stability of long-term effects and surrogate validity. While underscoring the irreplaceable value of high-quality long-term experimentation, the research also delineates core challenges and practical guidelines for surrogate learning in real-world settings.

0 citationsRead paper

Non-Exclusive Notifications for Ride-Hailing at Lyft II: Simulations and Marketplace Analysis

Mar 22, 2026

This study addresses the inefficiencies of traditional one-to-one exclusive ride dispatching, which often suffers from matching delays due to driver rejections or response timeouts, thereby reducing platform efficiency. The authors propose a non-exclusive dispatch mechanism that broadcasts each order to multiple drivers and develop an integrated evaluation framework combining a constrained welfare-maximization optimization model, large-scale discrete-event simulations based on real Lyft data, and macroscopic equilibrium analysis. Their findings reveal a fundamental trade-off between matching speed and service quality, demonstrating that non-exclusive dispatch significantly reduces matching time, decreases passenger abandonment, and improves both order completion rates and average service quality. Furthermore, they introduce a conservative notification strategy that mitigates over-allocation of high-value drivers, thereby enhancing long-term market efficiency.

0 citationsRead paper

Non-Exclusive Notifications for Ride-Hailing at Lyft I: Single-Cycle Approximation Algorithms

Mar 22, 2026

This study addresses the notification set selection problem for ride-hailing platforms under uncertainty in driver acceptance, aiming to maximize social welfare under non-exclusive notification mechanisms. For the “First-Accept” (FA) and “Best-Accept” (BA) protocols, the work presents the first PTAS and a constant-factor approximation algorithm, respectively. It establishes the monotone submodularity of the BA objective function and designs a demand oracle that surpasses the classical $1 - 1/e$ approximation barrier. Under homogeneous acceptance probabilities, an exact polynomial-time algorithm is further derived for the BA setting. Rigorous theoretical analysis provides strong approximation guarantees, while experiments on both synthetic data and real-world Lyft-calibrated datasets demonstrate the superior performance of the proposed algorithms in terms of matching quality and response efficiency.

0 citationsRead paper

Informal and Privatized Transit: Incentives, Efficiency and Coordination

Feb 11, 2026

This study addresses the critical yet often overlooked role of informal private transit in urban mobility, whose profit-driven operations are frequently neglected in planning, leading to systemic inefficiencies. The work proposes the first analytically tractable game-theoretic model of fully privatized informal transit, capturing strategic interactions between operators and cost-sensitive passengers. It introduces a tight bound on the Price of Anarchy and designs a Stackelberg routing framework coupled with route-specific cross-subsidization mechanisms. Theoretical analysis reveals that while decentralized decision-making incurs significant but bounded efficiency losses, minimal centralized intervention suffices to substantially mitigate them. Simulations based on real-world data from Nalasopara, India, empirically validate the effectiveness of the proposed mechanisms.

0 citationsRead paper

Post Launch Evaluation of Policies in a High-Dimensional Setting

Dec 30, 2024

In large-scale online platforms with hundreds of millions of users, conventional A/B testing is infeasible for post-hoc policy evaluation under sparse interventions, leading to severe estimation bias. To address this, we propose a two-stage debiased counterfactual estimation framework: (1) a covariate-based nearest-neighbor matching stage to construct high-fidelity control units and mitigate interpolation bias; and (2) a high-dimensional supervised learning stage—using XGBoost or neural networks—to model treatment effects while systematically diagnosing and correcting machine-learning-induced estimation bias. This is the first scalable solution enabling synthetic control methods to operate effectively under ultra-large-scale, sparse-intervention settings. Evaluated across six real-world online experiments, our method significantly improves causal effect estimation accuracy and reduces policy decision error rates by 42%. It has been deployed operationally to support closed-loop decision-making across multiple business units.

0 citationsRead paper