Akshit Kumar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Akshit Kumar has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
- Yash Kanoria
Graduate School of Business, Columbia University, New York, New York 10027
6 shared publications - Omar Besbes
Graduate School of Business, Columbia University, New York, New York 10027
4 shared publications - Wenxin Zhang
Graduate School of Business, Columbia University, New York, New York 10027
2 shared publications - Yilun Chen
School of Data Science, Chinese University of Hong Kong, Shenzhen 518172, China; and School of Management and Economics, Chinese University of Hong Kong, Shenzhen 518172, China
2 shared publications - Abdullah Althenayyan
Columbia University, New York, NY 10027, USA
1 shared publication - Akhilesh Tumu1 shared publication
Latest Publications
- Top Cited PapersOpen Access
- 4 September 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (36)
Digital twins are funhouse mirrors: Five systematic distortions
Scientists and practitioners are aggressively moving to deploy digital twins—large language model (LLM)-based models of individuals—across social science and policy research. We conducted 19 preregistered studies with 164 diverse outcomes (e.g., attitudes toward hiring algorithms and intention to share misinformation) and compared human responses with those of their digital twins (trained on each person’s previous answers to more than 500 questions). We establish an empirical benchmark for digital twin performance: Digital twins’ answers are only modestly more accurate than those from the (homogeneous) base LLM and correlate weakly with human responses (average correlation coefficient of 0.20). To guide future development, we document five ways in which digital twins distort human behavior: (i) insufficient individuation, (ii) stereotyping, (iii) representation bias, (iv) ideological biases, and (v) hyper-rationality. We make our full dataset and code public as a standardized testbed. Our results caution against premature deployment while laying the groundwork for the transparent, replicable, and iterative science necessary for responsible deployment of digital twins.
- 4 September 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (36)
Top Cited PapersOpen AccessDigital twins are funhouse mirrors: Five systematic distortions
Scientists and practitioners are aggressively moving to deploy digital twins—large language model (LLM)-based models of individuals—across social science and policy research. We conducted 19 preregistered studies with 164 diverse outcomes (e.g., attitudes toward hiring algorithms and intention to share misinformation) and compared human responses with those of their digital twins (trained on each person’s previous answers to more than 500 questions). We establish an empirical benchmark for digital twin performance: Digital twins’ answers are only modestly more accurate than those from the (homogeneous) base LLM and correlate weakly with human responses (average correlation coefficient of 0.20). To guide future development, we document five ways in which digital twins distort human behavior: (i) insufficient individuation, (ii) stereotyping, (iii) representation bias, (iv) ideological biases, and (v) hyper-rationality. We make our full dataset and code public as a standardized testbed. Our results caution against premature deployment while laying the groundwork for the transparent, replicable, and iterative science necessary for responsible deployment of digital twins.
- Open Access
- 11 June 2026
- Published by Elsevier in SSRN Electronic Journal
Feature-based Dynamic Matching in the Dark
We study a distribution-agnostic variant of the feature-based dynamic matching problem, where a set of supply units must be matched immediately and irrevocably to sequentially arriving demand units. Both sides are drawn i.i.d. from possibly different distributions over high-dimensional feature vectors, and the value of a match is a function of the two vectors. We measure performance by regret, the additive loss relative to the limiting hindsight-optimal match value. State-of-the-art algorithms for this problem require accurate knowledge of the demand distribution; we show that under model misspecification their performance can be highly sub-optimal. We develop ROAR, a completely distribution-agnostic algorithm that uses only the realized demand stream itself to make forward-looking assignment decisions. ROAR requires neither knowledge of the demand distribution nor any historical demand traces. We prove an exact identity expressing the average match value under ROAR as a folded average of hindsight-optimal values. As a consequence, ROAR obtains the same near-optimal regret scalings as the state-of-the-art algorithms that know the demand distribution exactly, across broad classes of quality functions and distributions.
- 11 June 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessFeature-based Dynamic Matching in the Dark
We study a distribution-agnostic variant of the feature-based dynamic matching problem, where a set of supply units must be matched immediately and irrevocably to sequentially arriving demand units. Both sides are drawn i.i.d. from possibly different distributions over high-dimensional feature vectors, and the value of a match is a function of the two vectors. We measure performance by regret, the additive loss relative to the limiting hindsight-optimal match value. State-of-the-art algorithms for this problem require accurate knowledge of the demand distribution; we show that under model misspecification their performance can be highly sub-optimal. We develop ROAR, a completely distribution-agnostic algorithm that uses only the realized demand stream itself to make forward-looking assignment decisions. ROAR requires neither knowledge of the demand distribution nor any historical demand traces. We prove an exact identity expressing the average match value under ROAR as a folded average of hindsight-optimal values. As a consequence, ROAR obtains the same near-optimal regret scalings as the state-of-the-art algorithms that know the demand distribution exactly, across broad classes of quality functions and distributions.
- Open Access
- 14 April 2026
- Published by Elsevier in SSRN Electronic Journal
How Much Should a Conversational Recommender System Converse?
Conversational recommender systems powered by generative AI can enhance personalization by facilitating information elicitation through follow-up questions. However, engaging in these conversations imposes a communication cost on users. As platforms with different objectives and monetization models deploy these systems, a central question is: how does the platform’s objective and sellers’ strategic response shape the design of these systems in terms of their elicitation strategy? We develop a parsimonious model of conversational elicitation in which interaction generates noisy preference information and imposes a communication cost borne by the user. A user-welfare-maximizing platform elicits more information when accurate niche matching yields large gains, even when niche users are rare. In contrast, under a conversion objective, for the same setting, the optimal strategy is to immediately recommend the same mainstream option to all users with no or minimal preference elicitation because the incremental conversion benefit from improved matching is bounded, while communication costs are borne by all users. When prices are endogenous and the platform earns a commission, increased elicitation is again optimal because improved screening raises equilibrium prices and platform revenue; however, these price responses can counteract consumer benefits and reduce user welfare. The model also highlights that the optimal elicitation intensity increases with preference heterogeneity, helping explain why conversational systems ask more in highly differentiated categories than in low-heterogeneity ones. We complement the theory with a dataset of long-form product queries that vary in length and informational content. Using our dataset and LLM-based user simulation, we quantify how additional information impacts user decisions and demonstrate that the magnitude of this impact depends on the degree of preference heterogeneity. Additionally, this dataset provides a testbed for measuring the (incremental) value of preference elicitation and may be of independent interest.
- 14 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessHow Much Should a Conversational Recommender System Converse?
Conversational recommender systems powered by generative AI can enhance personalization by facilitating information elicitation through follow-up questions. However, engaging in these conversations imposes a communication cost on users. As platforms with different objectives and monetization models deploy these systems, a central question is: how does the platform’s objective and sellers’ strategic response shape the design of these systems in terms of their elicitation strategy? We develop a parsimonious model of conversational elicitation in which interaction generates noisy preference information and imposes a communication cost borne by the user. A user-welfare-maximizing platform elicits more information when accurate niche matching yields large gains, even when niche users are rare. In contrast, under a conversion objective, for the same setting, the optimal strategy is to immediately recommend the same mainstream option to all users with no or minimal preference elicitation because the incremental conversion benefit from improved matching is bounded, while communication costs are borne by all users. When prices are endogenous and the platform earns a commission, increased elicitation is again optimal because improved screening raises equilibrium prices and platform revenue; however, these price responses can counteract consumer benefits and reduce user welfare. The model also highlights that the optimal elicitation intensity increases with preference heterogeneity, helping explain why conversational systems ask more in highly differentiated categories than in low-heterogeneity ones. We complement the theory with a dataset of long-form product queries that vary in length and informational content. Using our dataset and LLM-based user simulation, we quantify how additional information impacts user decisions and demonstrate that the magnitude of this impact depends on the degree of preference heterogeneity. Additionally, this dataset provides a testbed for measuring the (incremental) value of preference elicitation and may be of independent interest.
- Top Cited PapersOpen Access
- 12 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 8697-8700
What Is Your AI Agent Buying? Evaluation, Biases, Model Dependence, & Emerging Implications of Agentic E-Commerce
Online marketplaces will be transformed by autonomous AI agents acting on behalf of consumers. Rather than humans browsing and clicking, AI agents can parse webpages or interact through APIs to evaluate products, and transact. This raises a fundamental question: what do AI agents buy—and why? We develop ACES, a sandbox environment that pairs a platform-agnostic agent with a fully programmable mock marketplace to study this. We first explore aggregate choices, revealing that modal choices can differ across models, with AI agents sometimes concentrating on a few products, raising competition questions. We then analyze the current drivers of choices through randomized experiments on product positions and listing attributes. Models show sizeable and heterogeneous position effects: all favor the top row, yet different models prefer different columns, undermining the assumption of a universal ''top'' rank. They penalize sponsored tags, reward endorsements, and sensitivities to price, ratings, and reviews are directionally as expected, but vary sharply across models. Our findings reveal how AI agents behave in e-commerce, and surface concrete monitoring, seller strategy, platform design, and regulatory questions.
- 12 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 8697-8700
Top Cited PapersOpen AccessWhat Is Your AI Agent Buying? Evaluation, Biases, Model Dependence, & Emerging Implications of Agentic E-Commerce
Online marketplaces will be transformed by autonomous AI agents acting on behalf of consumers. Rather than humans browsing and clicking, AI agents can parse webpages or interact through APIs to evaluate products, and transact. This raises a fundamental question: what do AI agents buy—and why? We develop ACES, a sandbox environment that pairs a platform-agnostic agent with a fully programmable mock marketplace to study this. We first explore aggregate choices, revealing that modal choices can differ across models, with AI agents sometimes concentrating on a few products, raising competition questions. We then analyze the current drivers of choices through randomized experiments on product positions and listing attributes. Models show sizeable and heterogeneous position effects: all favor the top row, yet different models prefer different columns, undermining the assumption of a universal ''top'' rank. They penalize sponsored tags, reward endorsements, and sensitivities to price, ratings, and reviews are directionally as expected, but vary sharply across models. Our findings reveal how AI agents behave in e-commerce, and surface concrete monitoring, seller strategy, platform design, and regulatory questions.
- Top Cited Papers
- 1 March 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 74 (2) , 788-803
Feature-Based Dynamic Matching
SOAR ing to Optimality: Smarter Matching Algorithms for On-Demand Platforms In “Feature-Based Dynamic Matching,” Y. Chen, Y. Kanoria, A. Kumar, and W. Zhang study dynamic two-sided matching where both customers and service providers are characterized by high-dimensional feature vectors, motivated by platforms like on-demand home services. A key finding is that myopic greedy policies—which match each customer to the best available provider—can be highly suboptimal. The authors introduce SOAR (simulate-optimize-assign-repeat), a forward-looking algorithm that balances immediate match quality against preserving valuable supply for future arrivals. Through a novel analytical framework connecting dynamic matching to empirical optimal transport, they prove SOAR achieves near-optimal regret scaling under various distributional assumptions, establishing fundamental performance limits for feature-based matching problems and as a by-product resolve an open problem from prior work on dynamic spatial matching.
- 1 March 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 74 (2) , 788-803
Top Cited PapersFeature-Based Dynamic Matching
SOAR ing to Optimality: Smarter Matching Algorithms for On-Demand Platforms In “Feature-Based Dynamic Matching,” Y. Chen, Y. Kanoria, A. Kumar, and W. Zhang study dynamic two-sided matching where both customers and service providers are characterized by high-dimensional feature vectors, motivated by platforms like on-demand home services. A key finding is that myopic greedy policies—which match each customer to the best available provider—can be highly suboptimal. The authors introduce SOAR (simulate-optimize-assign-repeat), a forward-looking algorithm that balances immediate match quality against preserving valuable supply for future arrivals. Through a novel analytical framework connecting dynamic matching to empirical optimal transport, they prove SOAR achieves near-optimal regret scaling under various distributional assumptions, establishing fundamental performance limits for feature-based matching problems and as a by-product resolve an open problem from prior work on dynamic spatial matching.
- Open Access
- 2 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 352
Impact of Rankings and Personalized Recommendations in Marketplaces
- 2 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 352
Open AccessImpact of Rankings and Personalized Recommendations in Marketplaces
- 1 May 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 73 (3) , 1273-1288
Dynamic Resource Allocation: Algorithmic Design Principles and Spectrum of Achievable Performances
In “Dynamic Resource Allocation: Algorithmic Design Principles and Spectrum of Achievable Performances,” O. Besbes, Y. Kanoria, and A. Kumar consider a broad class of dynamic resource allocation problems and study the performance of practical algorithms. In particular, they focus on the interplay between the distribution of request types and achievable performance, given the broad set of configurations that can be encountered in practical settings. Although prior literature studied either a small number of request types or a continuum of types with no gaps, the present work allows for a large class of type distributions. Using initially the prototypical multisecretary problem to explore fundamental performance limits as a function of type distribution properties, the authors develop a new algorithmic property “conservativeness with respect to gaps” that guarantees near-optimal performance. In turn, they introduce a practically motivated, simulation-based algorithm and establish its near-optimal performance, not only for multisecretary problems, but also for general dynamic resource allocation problems.
- 1 May 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 73 (3) , 1273-1288
Dynamic Resource Allocation: Algorithmic Design Principles and Spectrum of Achievable Performances
In “Dynamic Resource Allocation: Algorithmic Design Principles and Spectrum of Achievable Performances,” O. Besbes, Y. Kanoria, and A. Kumar consider a broad class of dynamic resource allocation problems and study the performance of practical algorithms. In particular, they focus on the interplay between the distribution of request types and achievable performance, given the broad set of configurations that can be encountered in practical settings. Although prior literature studied either a small number of request types or a continuum of types with no gaps, the present work allows for a large class of type distributions. Using initially the prototypical multisecretary problem to explore fundamental performance limits as a function of type distribution properties, the authors develop a new algorithmic property “conservativeness with respect to gaps” that guarantees near-optimal performance. In turn, they introduce a practically motivated, simulation-based algorithm and establish its near-optimal performance, not only for multisecretary problems, but also for general dynamic resource allocation problems.
- Open Access
- 8 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 200-208
The Fault in Our Recommendations: On the Perils of Optimizing the Measurable
Recommendation systems are widespread, and through customized recommendations, promise to match users with options they will like. To that end, data on engagement is collected and used. Most recommendation systems are ranking-based, where they rank and recommend items based on their predicted engagement. However, the engagement signals are often only a crude proxy for user utility, as data on the latter is rarely collected or available. This paper explores the following question: By optimizing for measurable proxies, are recommendation systems at risk of significantly under-delivering on user utility? If that is indeed the case, how can one improve utility which is seldom measured? To study these questions, we introduce a model of repeated user consumption in which, at each interaction, users select between an outside option and the best option from a recommendation set. Our model accounts for user heterogeneity, with the majority preferring “popular” content, and a minority favoring “niche” content. The system initially lacks knowledge of individual user preferences but can learn these preferences through observations of users’ choices over time. Our theoretical and numerical analysis demonstrate that optimizing for engagement signals can lead to significant utility losses. Instead, we propose a utility-aware policy that initially recommends a mix of popular and niche content. We show that such a policy substantially improves utility despite not measuring it. As the platform becomes more forward-looking, our utility-aware policy achieves the best of both worlds: near-optimal user utility and near-optimal engagement simultaneously. Our study elucidates an important feature of recommendation systems; given the ability to suggest multiple items, one can perform significant exploration without incurring significant reductions in short term engagement. By recommending high-risk, high-reward items alongside popular items, systems can enhance discovery of high utility items without significantly affecting engagement.
- 8 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 200-208
Open AccessThe Fault in Our Recommendations: On the Perils of Optimizing the Measurable
Recommendation systems are widespread, and through customized recommendations, promise to match users with options they will like. To that end, data on engagement is collected and used. Most recommendation systems are ranking-based, where they rank and recommend items based on their predicted engagement. However, the engagement signals are often only a crude proxy for user utility, as data on the latter is rarely collected or available. This paper explores the following question: By optimizing for measurable proxies, are recommendation systems at risk of significantly under-delivering on user utility? If that is indeed the case, how can one improve utility which is seldom measured? To study these questions, we introduce a model of repeated user consumption in which, at each interaction, users select between an outside option and the best option from a recommendation set. Our model accounts for user heterogeneity, with the majority preferring “popular” content, and a minority favoring “niche” content. The system initially lacks knowledge of individual user preferences but can learn these preferences through observations of users’ choices over time. Our theoretical and numerical analysis demonstrate that optimizing for engagement signals can lead to significant utility losses. Instead, we propose a utility-aware policy that initially recommends a mix of popular and niche content. We show that such a policy substantially improves utility despite not measuring it. As the platform becomes more forward-looking, our utility-aware policy achieves the best of both worlds: near-optimal user utility and near-optimal engagement simultaneously. Our study elucidates an important feature of recommendation systems; given the ability to suggest multiple items, one can perform significant exploration without incurring significant reductions in short term engagement. By recommending high-risk, high-reward items alongside popular items, systems can enhance discovery of high utility items without significantly affecting engagement.
- Open Access
- 7 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 451
Feature Based Dynamic Matching
Motivated by matching platforms that match agents in a centralized manner, we introduce a model of dynamic two-sided matching where both demand and supply are heterogeneous with many types and the pool of supply units is limited. We model heterogeneity on the two sides of the market by i.i.d. demand weight vectors and i.i.d. supply feature vectors, with possibly different distributions. The matching of a demand-supply pair generates a utility that depends on their weight and feature vectors. To reflect the realistic structure of a heterogeneous matching market while also avoid impossibility results, we consider various levels of assumptions (in particular, the spatial structure) on matching utilities and feature distributions. The goal of the centralized platform is to dynamically assign supply units to sequentially arriving demand units in order to maximize utility. Many popular heuristic policies are either sub-optimal (like the myopic policy) or computationally inefficient (like the certainty equivalent policy). We propose a forward-looking supply-aware policy dubbed Simulate-Optimize-Assign-Repeat (SOAR) that combines practicality and strong theoretical guarantee. Inspired by model predictive control (MPC), SOAR leverages the power of simulation to balance between producing immediate high match utility and preserving valuable supply for future demands. We use regret as our performance metric for matching policies, specifically the additive loss relative to the utility per match achievable in the continuum limit (n → ∞). Under mild regularity assumptions on the offline matching instances, we prove that SOAR achieves the optimal regret scaling (up to a log factor). We further characterize the optimal regret scaling for interesting classes of problems with additional model structure, in particular, two classes of utility functions: (i) the "spatial utilities", namely the negative p-th power of the Euclidean distance between the supply and demand vectors where p ≥ 1; and (ii) the dot-product utility (equivalently p = 2 of (i)), and two classes of distributions: (i) both supply and demand distributions are smooth (a more stringent assumption) and (ii) supply and demand distributions are supported over compact sets (a mild assumption). En route to proving our guarantees we develop a novel framework for analyzing the performance of our SOAR policy which may be of wider applicability and independent interest. As a corollary of our techniques, we also resolve an open problem posed in Kanoria 2022.
- 7 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 451
Open AccessFeature Based Dynamic Matching
Motivated by matching platforms that match agents in a centralized manner, we introduce a model of dynamic two-sided matching where both demand and supply are heterogeneous with many types and the pool of supply units is limited. We model heterogeneity on the two sides of the market by i.i.d. demand weight vectors and i.i.d. supply feature vectors, with possibly different distributions. The matching of a demand-supply pair generates a utility that depends on their weight and feature vectors. To reflect the realistic structure of a heterogeneous matching market while also avoid impossibility results, we consider various levels of assumptions (in particular, the spatial structure) on matching utilities and feature distributions. The goal of the centralized platform is to dynamically assign supply units to sequentially arriving demand units in order to maximize utility. Many popular heuristic policies are either sub-optimal (like the myopic policy) or computationally inefficient (like the certainty equivalent policy). We propose a forward-looking supply-aware policy dubbed Simulate-Optimize-Assign-Repeat (SOAR) that combines practicality and strong theoretical guarantee. Inspired by model predictive control (MPC), SOAR leverages the power of simulation to balance between producing immediate high match utility and preserving valuable supply for future demands. We use regret as our performance metric for matching policies, specifically the additive loss relative to the utility per match achievable in the continuum limit (n → ∞). Under mild regularity assumptions on the offline matching instances, we prove that SOAR achieves the optimal regret scaling (up to a log factor). We further characterize the optimal regret scaling for interesting classes of problems with additional model structure, in particular, two classes of utility functions: (i) the "spatial utilities", namely the negative p-th power of the Euclidean distance between the supply and demand vectors where p ≥ 1; and (ii) the dot-product utility (equivalently p = 2 of (i)), and two classes of distributions: (i) both supply and demand distributions are smooth (a more stringent assumption) and (ii) supply and demand distributions are supported over compact sets (a mild assumption). En route to proving our guarantees we develop a novel framework for analyzing the performance of our SOAR policy which may be of wider applicability and independent interest. As a corollary of our techniques, we also resolve an open problem posed in Kanoria 2022.
- Open Access
- 1 January 2023
- Published by Elsevier in SSRN Electronic Journal
Feature Based Dynamic Matching
Motivated by matching platforms that match agents in a centralized manner, we study dynamic two-sided matching in a setting where both customers (demand) and service providers (supply) are heterogeneous and the pool of service providers is limited. We model heterogeneity on the two sides of the market by demand weight vectors drawn i.i.d. from some distribution, and supply feature vectors drawn i.i.d. from a (possibly) different distribution. The matching of a demand-supply pair generates a matching value that depends on their weight and feature vectors. We adopt a notion of regret, specifically the additive loss relative to the value (per match) achievable in the continuum limit, as our performance metric for matching policies. Simple myopic policies suffer non-vanishing regret in large markets. We propose a forward-looking supply-aware policy dubbed Simulate-Optimize-Assign-Repeat (SOAR) which balances between producing high match value for the current match and preserving valuable supply for future customers. We prove that SOAR achieves the optimal regret scaling under different assumptions on the demand and supply distributions. En-route to proving our guarantees we develop a novel framework for analyzing the performance of our SOAR policy which may be of broader interest. As a corollary of our techniques, we also resolve an open problem posed in Kanoria 2022.
- 1 January 2023
- Published by Elsevier in SSRN Electronic Journal
Open AccessFeature Based Dynamic Matching
Motivated by matching platforms that match agents in a centralized manner, we study dynamic two-sided matching in a setting where both customers (demand) and service providers (supply) are heterogeneous and the pool of service providers is limited. We model heterogeneity on the two sides of the market by demand weight vectors drawn i.i.d. from some distribution, and supply feature vectors drawn i.i.d. from a (possibly) different distribution. The matching of a demand-supply pair generates a matching value that depends on their weight and feature vectors. We adopt a notion of regret, specifically the additive loss relative to the value (per match) achievable in the continuum limit, as our performance metric for matching policies. Simple myopic policies suffer non-vanishing regret in large markets. We propose a forward-looking supply-aware policy dubbed Simulate-Optimize-Assign-Repeat (SOAR) which balances between producing high match value for the current match and preserving valuable supply for future customers. We prove that SOAR achieves the optimal regret scaling under different assumptions on the demand and supply distributions. En-route to proving our guarantees we develop a novel framework for analyzing the performance of our SOAR policy which may be of broader interest. As a corollary of our techniques, we also resolve an open problem posed in Kanoria 2022.