Lingzhou Xue
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
Lingzhou Xue has a total of 99 co-authors. Here are the top co-authors by number of shared publications:
Xiufan Yu
Department of Applied and Computational Mathematics and Statistics, University of Notre Dame , Notre Dame, IN 46556
7 shared publicationsHaoyi Yang
Department of Statistics, The Pennsylvania State University
6 shared publicationsAmal Agarwal
Department of Statistics, The Pennsylvania State University
5 shared publicationsKevin H. Lee
Department of Statistics, Western Michigan University
5 shared publications- Shiqian Ma
Department of Computational Applied Mathematics and Operations Research, Rice University, Houston, Texas 77005
5 shared publications - Xiang Zhan
Department of Biostatistics, School of Public Health, Peking University, Beijing, China
5 shared publications
Latest Publications
- Open Access
- 30 October 2026
- Published by MDPI AG in Preprints.org
From Harness Optimization to Recursive Self-Improvement: A Survey of Language Model Agents
Language-model agents rely on harnesses that govern tool use, context, memory, and execution. Understanding automated harness revision requires connecting improvement methods to the reusable functionality they produce. This survey examines three linked perspectives: improvement workflows, recursive scope, and concrete modifications across task domains. We compare search strategies, memory management, edit guidance, and integration with model training, and use an existing five-level framework to distinguish harness optimization from recursive self-improvement. Across coding, reasoning and information processing, interactive and embodied tasks, and visual tasks, we trace reported changes from starting capabilities to outcomes. Our synthesis identifies repairs at model–environment interfaces, cases where added tools and context fail to help, and conditions constraining recovery reuse. Cross-domain analogies motivate, rather than establish, transfer. We distinguish task-performance gains from improved subsequent optimization and identify priorities in reliable evaluation, attribution, transfer, and safety.
- 30 October 2026
- Published by MDPI AG in Preprints.org
Open AccessFrom Harness Optimization to Recursive Self-Improvement: A Survey of Language Model Agents
Language-model agents rely on harnesses that govern tool use, context, memory, and execution. Understanding automated harness revision requires connecting improvement methods to the reusable functionality they produce. This survey examines three linked perspectives: improvement workflows, recursive scope, and concrete modifications across task domains. We compare search strategies, memory management, edit guidance, and integration with model training, and use an existing five-level framework to distinguish harness optimization from recursive self-improvement. Across coding, reasoning and information processing, interactive and embodied tasks, and visual tasks, we trace reported changes from starting capabilities to outcomes. Our synthesis identifies repairs at model–environment interfaces, cases where added tools and context fail to help, and conditions constraining recovery reuse. Cross-domain analogies motivate, rather than establish, transfer. We distinguish task-performance gains from improved subsequent optimization and identify priorities in reliable evaluation, attribution, transfer, and safety.
- Open Access
- 20 October 2026
- Published by Taylor & Francis in Journal of the American Statistical Association
- p. 1-74
Variable Selection for Additive Global Fréchet Regression
- 20 October 2026
- Published by Taylor & Francis in Journal of the American Statistical Association
- p. 1-74
Open AccessVariable Selection for Additive Global Fréchet Regression
- Open Access
- 4 September 2026
- Published by Center for Open Science in PsyArXiv
Adaptive Regularization via Extreme Value Distributions for Gaussian Graphical Models
Edge selection in Gaussian graphical models is fundamentally a variable selection problem where pairwise relationships determine construct validity and variable importance in psychological networks. In psychology, network estimation relies predominantly on L1 regularization where uniform shrinkage systematically underestimates edge and centrality parameters. Alternative penalties overcome this bias but rely on fixed hyperparameters that do not adapt to the signal in the data. We develop a family of data-adaptive regularization penalties grounded in extreme value theory. Across 290 empirical psychological datasets, we show that absolute partial correlations are well-described by the Weibull distribution. Using this empirical regularity, we derive Weibull, Gumbel, and Exponential penalties that approximate L0 penalization and calibrate their hyperparameters to each dataset's noise floor. We formally prove the asymptotic properties of their static forms and conduct a large-scale simulation spanning two network topologies and various sample sizes (N = 100-10,000), demonstrating that their adaptive forms maintain high specificity while accumulating sensitivity as sample size increases with low parameter bias and high rank-order centrality congruence relative to field standards. Empirically, method choice alone determined centrality rankings at sample sizes typical in psychology. Of the three adaptive penalties, Weibull is recommended given the interpretability of its parameters.
- 4 September 2026
- Published by Center for Open Science in PsyArXiv
Open AccessAdaptive Regularization via Extreme Value Distributions for Gaussian Graphical Models
Edge selection in Gaussian graphical models is fundamentally a variable selection problem where pairwise relationships determine construct validity and variable importance in psychological networks. In psychology, network estimation relies predominantly on L1 regularization where uniform shrinkage systematically underestimates edge and centrality parameters. Alternative penalties overcome this bias but rely on fixed hyperparameters that do not adapt to the signal in the data. We develop a family of data-adaptive regularization penalties grounded in extreme value theory. Across 290 empirical psychological datasets, we show that absolute partial correlations are well-described by the Weibull distribution. Using this empirical regularity, we derive Weibull, Gumbel, and Exponential penalties that approximate L0 penalization and calibrate their hyperparameters to each dataset's noise floor. We formally prove the asymptotic properties of their static forms and conduct a large-scale simulation spanning two network topologies and various sample sizes (N = 100-10,000), demonstrating that their adaptive forms maintain high specificity while accumulating sensitivity as sample size increases with low parameter bias and high rank-order centrality congruence relative to field standards. Empirically, method choice alone determined centrality rankings at sample sizes typical in psychology. Of the three adaptive penalties, Weibull is recommended given the interpretability of its parameters.
- Open Access
Gap-Dependent Bounds for Nearly Minimax Optimal Reinforcement Learning with Linear Function Approximation
We study gap-dependent performance guarantees for nearly minimax-optimal algorithms in reinforcement learning with linear function approximation. While prior works have established gap-dependent regret bounds in this setting, existing analyses do not apply to algorithms that achieve the nearly minimax-optimal worst-case regret bound $\tilde{O}(d\sqrt{H^3K})$, where $d$ is the feature dimension, $H$ is the horizon length, and $K$ is the number of episodes. We bridge this gap by providing the first gap-dependent regret bound for the nearly minimax-optimal algorithm LSVI-UCB++ (He et al., 2023). Our analysis yields improved dependencies on both $d$ and $H$ compared to previous gap-dependent results. Moreover, leveraging the low policy-switching property of LSVI-UCB++, we introduce a concurrent variant that enables efficient parallel exploration across multiple agents and establish the first gap-dependent sample complexity upper bound for online multi-agent RL with linear function approximation, achieving linear speedup with respect to the number of agents.Open AccessGap-Dependent Bounds for Nearly Minimax Optimal Reinforcement Learning with Linear Function Approximation
We study gap-dependent performance guarantees for nearly minimax-optimal algorithms in reinforcement learning with linear function approximation. While prior works have established gap-dependent regret bounds in this setting, existing analyses do not apply to algorithms that achieve the nearly minimax-optimal worst-case regret bound $\tilde{O}(d\sqrt{H^3K})$, where $d$ is the feature dimension, $H$ is the horizon length, and $K$ is the number of episodes. We bridge this gap by providing the first gap-dependent regret bound for the nearly minimax-optimal algorithm LSVI-UCB++ (He et al., 2023). Our analysis yields improved dependencies on both $d$ and $H$ compared to previous gap-dependent results. Moreover, leveraging the low policy-switching property of LSVI-UCB++, we introduce a concurrent variant that enables efficient parallel exploration across multiple agents and establish the first gap-dependent sample complexity upper bound for online multi-agent RL with linear function approximation, achieving linear speedup with respect to the number of agents. - Open Access
Q-Learning with Fine-Grained Gap-Dependent Regret
We study fine-grained gap-dependent regret bounds for model-free reinforcement learning in episodic tabular Markov Decision Processes. Existing model-free algorithms achieve minimax worst-case regret, but their gap-dependent bounds remain coarse and fail to fully capture the structure of suboptimality gaps. We address this limitation by establishing fine-grained gap-dependent regret bounds for both UCB-based and non-UCB-based algorithms. In the UCB-based setting, we develop a novel analytical framework that explicitly separates the analysis of optimal and suboptimal state-action pairs, yielding the first fine-grained regret upper bound for UCB-Hoeffding (Jin et al., 2018). To highlight the generality of this framework, we introduce ULCB-Hoeffding, a new UCB-based algorithm inspired by AMB (Xu et al.,2021) but with a simplified structure, which enjoys fine-grained regret guarantees and empirically outperforms AMB. In the non-UCB-based setting, we revisit the only known algorithm AMB, and identify two key issues in its algorithm design and analysis: improper truncation in the $Q$-updates and violation of the martingale difference condition in its concentration argument. We propose a refined version of AMB that addresses these issues, establishing the first rigorous fine-grained gap-dependent regret for a non-UCB-based method, with experiments demonstrating improved performance over AMB.
Open AccessQ-Learning with Fine-Grained Gap-Dependent Regret
We study fine-grained gap-dependent regret bounds for model-free reinforcement learning in episodic tabular Markov Decision Processes. Existing model-free algorithms achieve minimax worst-case regret, but their gap-dependent bounds remain coarse and fail to fully capture the structure of suboptimality gaps. We address this limitation by establishing fine-grained gap-dependent regret bounds for both UCB-based and non-UCB-based algorithms. In the UCB-based setting, we develop a novel analytical framework that explicitly separates the analysis of optimal and suboptimal state-action pairs, yielding the first fine-grained regret upper bound for UCB-Hoeffding (Jin et al., 2018). To highlight the generality of this framework, we introduce ULCB-Hoeffding, a new UCB-based algorithm inspired by AMB (Xu et al.,2021) but with a simplified structure, which enjoys fine-grained regret guarantees and empirically outperforms AMB. In the non-UCB-based setting, we revisit the only known algorithm AMB, and identify two key issues in its algorithm design and analysis: improper truncation in the $Q$-updates and violation of the martingale difference condition in its concentration argument. We propose a refined version of AMB that addresses these issues, establishing the first rigorous fine-grained gap-dependent regret for a non-UCB-based method, with experiments demonstrating improved performance over AMB.
- 1 September 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 71  (9) , 7312-7332
A Graphical Point Process Framework for Understanding Removal Effects in Multi-Touch Attribution
Marketers rely on various online advertising channels to reach customers and are increasingly interested in multi-touch attribution, which evaluates the contribution of each touchpoint to a conversion. However, as the numbers of marketing channels and touchpoints increase, the attribution challenge becomes more intricate because of the complex interplay among different touchpoints within and across channels. Utilizing customer path-to-purchase data, this article addresses this challenge by developing a novel graphical point process framework to investigate the relational structure among various touchpoints. Based on this framework, we propose graphical attribution methods that allocate attribution scores to individual touchpoints or corresponding channels for each customer’s path to purchase. These scores are calculated using a probabilistic definition of removal effects. We evaluate the proposed methods and compare their performance with commonly used attribution models through extensive simulation studies and a real-world attribution application. This paper was accepted by J. George Shanthikumar, data science. Funding: This work was supported by the National Science Foundation [Grant DMS-2210775]. Additionally, the research of L. Xue was partly supported by the National Science Foundation [Grants DMS-1811552 and DMS-1953189]. Supplemental Material: The electronic companion and data files are available at https://doi.org/10.1287/mnsc.2023.00457 .
- 1 September 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
- Vol. 71  (9) , 7312-7332
A Graphical Point Process Framework for Understanding Removal Effects in Multi-Touch Attribution
Marketers rely on various online advertising channels to reach customers and are increasingly interested in multi-touch attribution, which evaluates the contribution of each touchpoint to a conversion. However, as the numbers of marketing channels and touchpoints increase, the attribution challenge becomes more intricate because of the complex interplay among different touchpoints within and across channels. Utilizing customer path-to-purchase data, this article addresses this challenge by developing a novel graphical point process framework to investigate the relational structure among various touchpoints. Based on this framework, we propose graphical attribution methods that allocate attribution scores to individual touchpoints or corresponding channels for each customer’s path to purchase. These scores are calculated using a probabilistic definition of removal effects. We evaluate the proposed methods and compare their performance with commonly used attribution models through extensive simulation studies and a real-world attribution application. This paper was accepted by J. George Shanthikumar, data science. Funding: This work was supported by the National Science Foundation [Grant DMS-2210775]. Additionally, the research of L. Xue was partly supported by the National Science Foundation [Grants DMS-1811552 and DMS-1953189]. Supplemental Material: The electronic companion and data files are available at https://doi.org/10.1287/mnsc.2023.00457 .
- Open Access
- 11 August 2025
- Published by Springer Nature in Genome Biology
- Vol. 26  (1)
spVelo: RNA velocity inference for multi-batch spatial transcriptomics data
RNA velocity has emerged as a powerful tool to interpret transcriptional dynamics and infer trajectory from snapshot datasets. However, current methods fail to utilize the spatial information inherent in spatial transcriptomics and lack scalability in multi-batch datasets. Here, we introduce spVelo, a scalable framework for RNA velocity inference of multi-batch spatial transcriptomics data. spVelo supports several downstream applications, including uncertainty quantification, complex trajectory pattern discovery, driver marker identification, gene regulatory network inference, and temporal cell-cell communication inference. spVelo has the potential to provide deeper insights into complex tissue organization and underscore biological mechanisms based on spatially resolved patterns.
- 11 August 2025
- Published by Springer Nature in Genome Biology
- Vol. 26  (1)
Open AccessspVelo: RNA velocity inference for multi-batch spatial transcriptomics data
RNA velocity has emerged as a powerful tool to interpret transcriptional dynamics and infer trajectory from snapshot datasets. However, current methods fail to utilize the spatial information inherent in spatial transcriptomics and lack scalability in multi-batch datasets. Here, we introduce spVelo, a scalable framework for RNA velocity inference of multi-batch spatial transcriptomics data. spVelo supports several downstream applications, including uncertainty quantification, complex trajectory pattern discovery, driver marker identification, gene regulatory network inference, and temporal cell-cell communication inference. spVelo has the potential to provide deeper insights into complex tissue organization and underscore biological mechanisms based on spatially resolved patterns.
- Open Access
- 1 July 2025
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 41  (7)
Composition-on-composition regression analysis for multi-omics integration of metagenomic data
Motivation: Compositional data are frequently encountered in many disciplines, such as in next-generation sequencing experiments widely used in biomedical studies. Regression analysis with compositional data as either responses or predictors has been well studied. However, when both responses and predictors are compositional, the inventory of analysis tools is surprisingly limited, especially in the high-dimensional setting. Among the few existing methods, most of them rely on a log-ratio transformation to move compositional data from the simplex to real numbers. Yet, a serious weakness of these methods is their failure to handle the substantial fraction of zeroes observed in data collected from next-generation sequencing experiments. Results: To investigate associations between two high-dimensional multi-omics compositions, we propose a composition-on-composition (COC) regression analysis method which does not require log-ratio transformations and hence can handle zeroes in the data. To account for high dimensionality, we estimate regression coefficients using a penalized estimation equation approach. Finally, inference procedures for COC regression are also proposed. Superior performance of COC is demonstrated through both comprehensive numerical simulations and case studies. Availability and implementation: Source R codes to implement COC method is available at https://github.com/nrios4/COC.
- 1 July 2025
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 41  (7)
Open AccessComposition-on-composition regression analysis for multi-omics integration of metagenomic data
Motivation: Compositional data are frequently encountered in many disciplines, such as in next-generation sequencing experiments widely used in biomedical studies. Regression analysis with compositional data as either responses or predictors has been well studied. However, when both responses and predictors are compositional, the inventory of analysis tools is surprisingly limited, especially in the high-dimensional setting. Among the few existing methods, most of them rely on a log-ratio transformation to move compositional data from the simplex to real numbers. Yet, a serious weakness of these methods is their failure to handle the substantial fraction of zeroes observed in data collected from next-generation sequencing experiments. Results: To investigate associations between two high-dimensional multi-omics compositions, we propose a composition-on-composition (COC) regression analysis method which does not require log-ratio transformations and hence can handle zeroes in the data. To account for high dimensionality, we estimate regression coefficients using a penalized estimation equation approach. Finally, inference procedures for COC regression are also proposed. Superior performance of COC is demonstrated through both comprehensive numerical simulations and case studies. Availability and implementation: Source R codes to implement COC method is available at https://github.com/nrios4/COC.
- 23 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Theory
- Vol. 71  (9) , 7149-7175
Theoretical Guarantees for Sparse Principal Component Analysis Based on the Elastic Net
Sparse principal component analysis (SPCA) is widely used for dimensionality reduction and feature extraction in high-dimensional data analysis. Despite many methodological and theoretical developments in the past two decades, the theoretical guarantees of the popular SPCA algorithm proposed by Zou et al. (2006) based on the elastic net are still unknown. This paper aims to address this critical theoretical gap. We first revisit the SPCA algorithm of Zou et al. (2006) and present our implementation. We also study a computationally more efficient variant of the SPCA algorithm in Zou et al. (2006) that can be considered as the limiting case of SPCA. We provide the guarantees of convergence to a stationary point for both algorithms and prove that, under a sparse spiked covariance model, both algorithms can recover the principal subspace consistently under mild regularity conditions. We show that their estimation error bounds match the best available bounds of existing works or the minimax rates up to some logarithmic factors. Moreover, we demonstrate the competitive numerical performance of both algorithms in numerical experiments.
- 23 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Theory
- Vol. 71  (9) , 7149-7175
Theoretical Guarantees for Sparse Principal Component Analysis Based on the Elastic Net
Sparse principal component analysis (SPCA) is widely used for dimensionality reduction and feature extraction in high-dimensional data analysis. Despite many methodological and theoretical developments in the past two decades, the theoretical guarantees of the popular SPCA algorithm proposed by Zou et al. (2006) based on the elastic net are still unknown. This paper aims to address this critical theoretical gap. We first revisit the SPCA algorithm of Zou et al. (2006) and present our implementation. We also study a computationally more efficient variant of the SPCA algorithm in Zou et al. (2006) that can be considered as the limiting case of SPCA. We provide the guarantees of convergence to a stationary point for both algorithms and prove that, under a sparse spiked covariance model, both algorithms can recover the principal subspace consistently under mild regularity conditions. We show that their estimation error bounds match the best available bounds of existing works or the minimax rates up to some logarithmic factors. Moreover, we demonstrate the competitive numerical performance of both algorithms in numerical experiments.
- Open Access
- 20 May 2025
- Published by Taylor & Francis in Journal of Computational and Graphical Statistics
- Vol. 34  (4) , 1692-1704
Clustering Time-Evolving Networks Using Temporal Exponential-Family Random Graph Models with Conditional Dyadic Independence and Dynamic Latent Blocks
- 20 May 2025
- Published by Taylor & Francis in Journal of Computational and Graphical Statistics
- Vol. 34  (4) , 1692-1704
Open AccessClustering Time-Evolving Networks Using Temporal Exponential-Family Random Graph Models with Conditional Dyadic Independence and Dynamic Latent Blocks