Kuang Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Kuang Xu has a total of 116 co-authors. Here are the top co-authors by number of shared publications:
- Stefan Wager
Stanford Graduate School of Business, Stanford University
6 shared publications - John N. Tsitsiklis
Laboratory for Information and Decision Systems, Electrical Engineering and Computer Science Department, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139
4 shared publications - Benjamin Van Roy
Stanford University Stanford CA
3 shared publications Dana Yang
2 shared publicationsJiaming Xu
2 shared publications- Ramesh Johari
Stanford University, Stanford, CA, USA
2 shared publications
Latest Publications
- Open Access
- 3 November 2026
- Published by MDPI AG in Preprints.org
Agentic Laboratories of the Future: Towards World Models for Scientific Discovery
Advances in scientific AI have accelerated hypothesis generation, prediction, and experimental design, but translating these capabilities into reliable physical experimentation remains a major bottleneck in scientific discovery. Although automation, high-throughput experimentation, foundation models, and cloud infrastructure have accelerated individual stages of the scientific workflow, they have not unified the discovery process. Scientists must still coordinate between computational models and physical laboratory systems to translate scientific designs into executable experiments and ensure their results are reliable. We argue that the next generation of laboratories will be agentic: environments in which scientists, AI systems, equipment and robotic platforms operate as collaborative discovery partners. We propose an architecture built around two complementary components: an agentic foundation that translates scientific designs into coordinated execution plans, and an evolving laboratory world model that represents physical state, predicts action outcomes, and incorporates experimental feedback. Together, they connect computational reasoning to observable and verifiable physical execution. Multimodal perception, robotics, and human–AI interfaces can make laboratory actions and evolving experimental conditions accessible to these systems, supporting adaptive execution and learning. We further propose an evaluation framework centered on physical grounding, execution fidelity, scientific validity, and adaptive coordination under human oversight, rather than autonomy alone. This Perspective outlines how agentic laboratories could connect scientific reasoning, physical experimentation, and trustworthy evidence within a human-in-the-lead discovery loop.
- 3 November 2026
- Published by MDPI AG in Preprints.org
Open AccessAgentic Laboratories of the Future: Towards World Models for Scientific Discovery
Advances in scientific AI have accelerated hypothesis generation, prediction, and experimental design, but translating these capabilities into reliable physical experimentation remains a major bottleneck in scientific discovery. Although automation, high-throughput experimentation, foundation models, and cloud infrastructure have accelerated individual stages of the scientific workflow, they have not unified the discovery process. Scientists must still coordinate between computational models and physical laboratory systems to translate scientific designs into executable experiments and ensure their results are reliable. We argue that the next generation of laboratories will be agentic: environments in which scientists, AI systems, equipment and robotic platforms operate as collaborative discovery partners. We propose an architecture built around two complementary components: an agentic foundation that translates scientific designs into coordinated execution plans, and an evolving laboratory world model that represents physical state, predicts action outcomes, and incorporates experimental feedback. Together, they connect computational reasoning to observable and verifiable physical execution. Multimodal perception, robotics, and human–AI interfaces can make laboratory actions and evolving experimental conditions accessible to these systems, supporting adaptive execution and learning. We further propose an evaluation framework centered on physical grounding, execution fidelity, scientific validity, and adaptive coordination under human oversight, rather than autonomy alone. This Perspective outlines how agentic laboratories could connect scientific reasoning, physical experimentation, and trustworthy evidence within a human-in-the-lead discovery loop.
- 10 August 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
Experimenting Under Stochastic Congestion
We study randomized experiments in a service system when stochastic congestion can arise from temporarily limited supply or excess demand. Such congestion gives rise to cross-unit interference between the waiting customers, and analytic strategies that do not account for this interference may be biased. In current practice, one of the most widely used ways to address stochastic congestion is to use switchback experiments that turn a target intervention on and off for the whole system in alternation. We find, however, that under a queueing model for stochastic congestion, the standard way of analyzing switchbacks is inefficient and that estimators that leverage the queueing model can be materially more accurate. Additionally, we show how the queueing model enables estimation of total policy gradients from unit-level randomized experiments, thus giving practitioners an alternative experimental approach they can use without needing to precommit to a fixed switchback length before data collection. This paper was accepted by Vivek Farias, data science. Funding: This work was supported by the ONR [Grant N000142412091]. Supplemental Material: The online appendices and data files are available at https://doi.org/10.1287/mnsc.2024.08306 .
- 10 August 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
Experimenting Under Stochastic Congestion
We study randomized experiments in a service system when stochastic congestion can arise from temporarily limited supply or excess demand. Such congestion gives rise to cross-unit interference between the waiting customers, and analytic strategies that do not account for this interference may be biased. In current practice, one of the most widely used ways to address stochastic congestion is to use switchback experiments that turn a target intervention on and off for the whole system in alternation. We find, however, that under a queueing model for stochastic congestion, the standard way of analyzing switchbacks is inefficient and that estimators that leverage the queueing model can be materially more accurate. Additionally, we show how the queueing model enables estimation of total policy gradients from unit-level randomized experiments, thus giving practitioners an alternative experimental approach they can use without needing to precommit to a fixed switchback length before data collection. This paper was accepted by Vivek Farias, data science. Funding: This work was supported by the ONR [Grant N000142412091]. Supplemental Material: The online appendices and data files are available at https://doi.org/10.1287/mnsc.2024.08306 .
- 1 March 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 74 (2) , 1026-1046
Load Balancing Using Sparse Communication
Smarter Load Balancing with Fewer Messages Modern data centers rely on real-time information to route jobs efficiently—but constant communication between servers and load balancers can overwhelm the network. In their paper, “Load Balancing Using Sparse Communication,” Gal Mendelson and Kuang Xu present a breakthrough: high-performance load balancing using drastically fewer messages. They introduce a flexible framework based on state approximation and develop new algorithms and communication protocols that maintain near-optimal performance even when communication is sparse. The key insight is that servers can monitor how wrong the load balancer’s estimate is and communicate only when necessary. Their approach reduces communication by over 90% while sacrificing little quality, as proven through both theory and simulation. These results offer a scalable and practical solution for large service systems where bandwidth is precious.
- 1 March 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 74 (2) , 1026-1046
Load Balancing Using Sparse Communication
Smarter Load Balancing with Fewer Messages Modern data centers rely on real-time information to route jobs efficiently—but constant communication between servers and load balancers can overwhelm the network. In their paper, “Load Balancing Using Sparse Communication,” Gal Mendelson and Kuang Xu present a breakthrough: high-performance load balancing using drastically fewer messages. They introduce a flexible framework based on state approximation and develop new algorithms and communication protocols that maintain near-optimal performance even when communication is sparse. The key insight is that servers can monitor how wrong the load balancer’s estimate is and communicate only when necessary. Their approach reduces communication by over 90% while sacrificing little quality, as proven through both theory and simulation. These results offer a scalable and practical solution for large service systems where bandwidth is precious.
- 1 October 2024
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 70 (10) , 7024-7041
Weak Signal Asymptotics for Sequentially Randomized Experiments
We use the lens of weak signal asymptotics to study a class of sequentially randomized experiments, including those that arise in solving multiarmed bandit problems. In an experiment with n time steps, we let the mean reward gaps between actions scale to the order [ ] to preserve the difficulty of the learning task as n grows. In this regime, we show that the sample paths of a class of sequentially randomized experiments—adapted to this scaling regime and with arm selection probabilities that vary continuously with state—converge weakly to a diffusion limit, given as the solution to a stochastic differential equation. The diffusion limit enables us to derive refined, instance-specific characterization of stochastic dynamics and to obtain several insights on the regret and belief evolution of a number of sequential experiments including Thompson sampling (but not upper-confidence bound, which does not satisfy our continuity assumption). We show that all sequential experiments whose randomization probabilities have a Lipschitz-continuous dependence on the observed data suffer from suboptimal regret performance when the reward gaps are relatively large. Conversely, we find that a version of Thompson sampling with an asymptotically uninformative prior variance achieves near-optimal instance-specific regret scaling, including with large reward gaps, but these good regret properties come at the cost of highly unstable posterior beliefs. This paper was accepted by Baris Ata, stochastic models and simulation. Supplemental Material: The data and online appendix are available at https://doi.org/10.1287/mnsc.2023.4964 .
- 1 October 2024
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 70 (10) , 7024-7041
Weak Signal Asymptotics for Sequentially Randomized Experiments
We use the lens of weak signal asymptotics to study a class of sequentially randomized experiments, including those that arise in solving multiarmed bandit problems. In an experiment with n time steps, we let the mean reward gaps between actions scale to the order [ ] to preserve the difficulty of the learning task as n grows. In this regime, we show that the sample paths of a class of sequentially randomized experiments—adapted to this scaling regime and with arm selection probabilities that vary continuously with state—converge weakly to a diffusion limit, given as the solution to a stochastic differential equation. The diffusion limit enables us to derive refined, instance-specific characterization of stochastic dynamics and to obtain several insights on the regret and belief evolution of a number of sequential experiments including Thompson sampling (but not upper-confidence bound, which does not satisfy our continuity assumption). We show that all sequential experiments whose randomization probabilities have a Lipschitz-continuous dependence on the observed data suffer from suboptimal regret performance when the reward gaps are relatively large. Conversely, we find that a version of Thompson sampling with an asymptotically uninformative prior variance achieves near-optimal instance-specific regret scaling, including with large reward gaps, but these good regret properties come at the cost of highly unstable posterior beliefs. This paper was accepted by Baris Ata, stochastic models and simulation. Supplemental Material: The data and online appendix are available at https://doi.org/10.1287/mnsc.2023.4964 .
- Open Access
- 8 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 780
Experimenting under Stochastic Congestion
- 8 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 780
Open AccessExperimenting under Stochastic Congestion
- Open Access
Non-Stationary Contextual Bandit Learning via Neural Predictive Ensemble Sampling
Real-world applications of contextual bandits often exhibit non-stationarity due to seasonality, serendipity, and evolving social trends. While a number of non-stationary contextual bandit learning algorithms have been proposed in the literature, they excessively explore due to a lack of prioritization for information of enduring value, or are designed in ways that do not scale in modern applications with high-dimensional user-specific features and large action set, or both. In this paper, we introduce a novel non-stationary contextual bandit algorithm that addresses these concerns. It combines a scalable, deep-neural-network-based architecture with a carefully designed exploration mechanism that strategically prioritizes collecting information with the most lasting value in a non-stationary environment. Through empirical evaluations on two real-world recommendation datasets, which exhibit pronounced non-stationarity, we demonstrate that our approach significantly outperforms the state-of-the-art baselines.
Open AccessNon-Stationary Contextual Bandit Learning via Neural Predictive Ensemble Sampling
Real-world applications of contextual bandits often exhibit non-stationarity due to seasonality, serendipity, and evolving social trends. While a number of non-stationary contextual bandit learning algorithms have been proposed in the literature, they excessively explore due to a lack of prioritization for information of enduring value, or are designed in ways that do not scale in modern applications with high-dimensional user-specific features and large action set, or both. In this paper, we introduce a novel non-stationary contextual bandit algorithm that addresses these concerns. It combines a scalable, deep-neural-network-based architecture with a carefully designed exploration mechanism that strategically prioritizes collecting information with the most lasting value in a non-stationary environment. Through empirical evaluations on two real-world recommendation datasets, which exhibit pronounced non-stationarity, we demonstrate that our approach significantly outperforms the state-of-the-art baselines.
- Open Access
Non-Stationary Bandit Learning via Predictive Sampling
Thompson sampling has proven effective across a wide range of stationary bandit environments. However, as we demonstrate in this paper, it can perform poorly when applied to non-stationary environments. We attribute such failures to the fact that, when exploring, the algorithm does not differentiate actions based on how quickly the information acquired loses its usefulness due to non-stationarity. Building upon this insight, we propose predictive sampling, an algorithm that deprioritizes acquiring information that quickly loses usefulness. A theoretical guarantee on the performance of predictive sampling is established through a Bayesian regret bound. We provide versions of predictive sampling for which computations tractably scale to complex bandit environments of practical interest. Through numerical simulations, we demonstrate that predictive sampling outperforms Thompson sampling in all non-stationary environments examined.
Open AccessNon-Stationary Bandit Learning via Predictive Sampling
Thompson sampling has proven effective across a wide range of stationary bandit environments. However, as we demonstrate in this paper, it can perform poorly when applied to non-stationary environments. We attribute such failures to the fact that, when exploring, the algorithm does not differentiate actions based on how quickly the information acquired loses its usefulness due to non-stationarity. Building upon this insight, we propose predictive sampling, an algorithm that deprioritizes acquiring information that quickly loses usefulness. A theoretical guarantee on the performance of predictive sampling is established through a Bayesian regret bound. We provide versions of predictive sampling for which computations tractably scale to complex bandit environments of practical interest. Through numerical simulations, we demonstrate that predictive sampling outperforms Thompson sampling in all non-stationary environments examined.
- Open Access
Weak Signal Asymptotics for Sequentially Randomized Experiments
We use the lens of weak signal asymptotics to study a class of sequentially randomized experiments, including those that arise in solving multi-armed bandit problems. In an experiment with $n$ time steps, we let the mean reward gaps between actions scale to the order $1/\sqrt{n}$ so as to preserve the difficulty of the learning task as $n$ grows. In this regime, we show that the sample paths of a class of sequentially randomized experiments -- adapted to this scaling regime and with arm selection probabilities that vary continuously with state -- converge weakly to a diffusion limit, given as the solution to a stochastic differential equation. The diffusion limit enables us to derive refined, instance-specific characterization of stochastic dynamics, and to obtain several insights on the regret and belief evolution of a number of sequential experiments including Thompson sampling (but not UCB, which does not satisfy our continuity assumption). We show that all sequential experiments whose randomization probabilities have a Lipschitz-continuous dependence on the observed data suffer from sub-optimal regret performance when the reward gaps are relatively large. Conversely, we find that a version of Thompson sampling with an asymptotically uninformative prior variance achieves near-optimal instance-specific regret scaling, including with large reward gaps, but these good regret properties come at the cost of highly unstable posterior beliefs.Open AccessWeak Signal Asymptotics for Sequentially Randomized Experiments
We use the lens of weak signal asymptotics to study a class of sequentially randomized experiments, including those that arise in solving multi-armed bandit problems. In an experiment with $n$ time steps, we let the mean reward gaps between actions scale to the order $1/\sqrt{n}$ so as to preserve the difficulty of the learning task as $n$ grows. In this regime, we show that the sample paths of a class of sequentially randomized experiments -- adapted to this scaling regime and with arm selection probabilities that vary continuously with state -- converge weakly to a diffusion limit, given as the solution to a stochastic differential equation. The diffusion limit enables us to derive refined, instance-specific characterization of stochastic dynamics, and to obtain several insights on the regret and belief evolution of a number of sequential experiments including Thompson sampling (but not UCB, which does not satisfy our continuity assumption). We show that all sequential experiments whose randomization probabilities have a Lipschitz-continuous dependence on the observed data suffer from sub-optimal regret performance when the reward gaps are relatively large. Conversely, we find that a version of Thompson sampling with an asymptotically uninformative prior variance achieves near-optimal instance-specific regret scaling, including with large reward gaps, but these good regret properties come at the cost of highly unstable posterior beliefs. - 5 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Theory
- Vol. 69 (1) , 528-547
Learner-Private Convex Optimization
Convex optimization with feedback is a framework where a learner relies on iterative queries and feedback to arrive at the minimizer of a convex function. It has gained considerable popularity thanks to its scalability in large-scale optimization and machine learning. The repeated interactions, however, expose the learner to privacy risks from eavesdropping adversaries that observe the submitted queries. In this paper, we study how to optimally obfuscate the learner’s queries in convex optimization with first-order feedback, so that their learned optimal value is provably difficult to estimate for an eavesdropping adversary. We consider two formulations of learner privacy: a Bayesian formulation in which the convex function is drawn randomly, and a maximin formulation in which the function is fixed and the adversary’s probability of error is measured with respect to a minimax criterion. Suppose that the learner wishes to ensure the adversary cannot estimate accurately with probability greater than 1/ L for some L > 0. Our main results show that the query complexity overhead is additive in L in the maximin formulation, but multiplicative in L in the Bayesian formulation. Compared to existing learner-private sequential learning models with binary feedback, our results apply to the significantly richer family of general convex functions with full-gradient feedback. Our proofs rely on tools from the theory of Dirichlet processes, as well as a novel strategy designed for measuring information leakage under a full-gradient oracle.
- 5 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Theory
- Vol. 69 (1) , 528-547
Learner-Private Convex Optimization
Convex optimization with feedback is a framework where a learner relies on iterative queries and feedback to arrive at the minimizer of a convex function. It has gained considerable popularity thanks to its scalability in large-scale optimization and machine learning. The repeated interactions, however, expose the learner to privacy risks from eavesdropping adversaries that observe the submitted queries. In this paper, we study how to optimally obfuscate the learner’s queries in convex optimization with first-order feedback, so that their learned optimal value is provably difficult to estimate for an eavesdropping adversary. We consider two formulations of learner privacy: a Bayesian formulation in which the convex function is drawn randomly, and a maximin formulation in which the function is fixed and the adversary’s probability of error is measured with respect to a minimax criterion. Suppose that the learner wishes to ensure the adversary cannot estimate accurately with probability greater than 1/ L for some L > 0. Our main results show that the query complexity overhead is additive in L in the maximin formulation, but multiplicative in L in the Bayesian formulation. Compared to existing learner-private sequential learning models with binary feedback, our results apply to the significantly richer family of general convex functions with full-gradient feedback. Our proofs rely on tools from the theory of Dirichlet processes, as well as a novel strategy designed for measuring information leakage under a full-gradient oracle.
- Open Access
Temporal Concatenation for Markov Decision Processes
We propose and analyze a temporal concatenation heuristic for solving large-scale finite-horizon Markov decision processes (MDP), which divides the MDP into smaller sub-problems along the time horizon and generates an overall solution by simply concatenating the optimal solutions from these sub-problems. As a "black box" architecture, temporal concatenation works with a wide range of existing MDP algorithms. Our main results characterize the regret of temporal concatenation compared to the optimal solution. We provide upper bounds for general MDP instances, as well as a family of MDP instances in which the upper bounds are shown to be tight. Together, our results demonstrate temporal concatenation's potential of substantial speed-up at the expense of some performance degradation.
Open AccessTemporal Concatenation for Markov Decision Processes
We propose and analyze a temporal concatenation heuristic for solving large-scale finite-horizon Markov decision processes (MDP), which divides the MDP into smaller sub-problems along the time horizon and generates an overall solution by simply concatenating the optimal solutions from these sub-problems. As a "black box" architecture, temporal concatenation works with a wide range of existing MDP algorithms. Our main results characterize the regret of temporal concatenation compared to the optimal solution. We provide upper bounds for general MDP instances, as well as a family of MDP instances in which the upper bounds are shown to be tight. Together, our results demonstrate temporal concatenation's potential of substantial speed-up at the expense of some performance degradation.