Haijun Yang
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
Haijun Yang has a total of 75 co-authors. Here are the top co-authors by number of shared publications:
Shancun Liu
School of Economics and Management, Beihang University, Beijing, China
6 shared publicationsHarris Wu
Strome College of Business Old Dominion University Norfolk Virginia USA
5 shared publicationsHui Zhang
4 shared publications- Zhenyuan Huang
Beihang University
4 shared publications - Harris Wu
Strome College of Business, Old Dominion University, Norfolk, Virginia 23529
3 shared publications - Hui Zhang
Department of Computer Science and Engineering, Beihang University, Beijing, China
3 shared publications
Latest Publications
- 30 October 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 355
SPARAG: Differentially private Retrieval-Augmented Generation via sampling and aggregation
- 30 October 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 355
SPARAG: Differentially private Retrieval-Augmented Generation via sampling and aggregation
- 29 October 2026
- Published by Elsevier in International Review of Financial Analysis
- Vol. 119
Strategic evolution in cryptocurrency market: A tripartite evolutionary game perspective
- 29 October 2026
- Published by Elsevier in International Review of Financial Analysis
- Vol. 119
Strategic evolution in cryptocurrency market: A tripartite evolutionary game perspective
- Open Access
- 28 April 2026
- Published by Elsevier in SSRN Electronic Journal
The optimal mechanism in high-frequency batch trading with multiple news
We develop a new high-frequency batch model based on game theory to analyze market behavior and address the issue of optimal trading strategies under multiple news arrival scenarios. In this framework, the behaviors and the profits of high-frequency traders (HFTs) are critically influenced by both the quantity and the sequence of news arrival. The model incorporates three interconnected games among HFTs and is refined to identify the trembling-hand perfect Nash equilibrium (THPNE). By maximizing the total market welfare measure, an optimal batch interval is determined to achieve the most efficient THPNE within the batch mechanism. Within this equilibrium, we characterize the optimal spread strategy and dynamic order strategy of HFTs. Single- and multi-period batch models are constructed to compute payoff matrices by analyzing market microstructure and HFTs’ behavior across multiple sequential news arrival scenarios, thereby enabling the derivation of the market spread and HFT profits. Finally, we show the optimal batch interval, a key practical finding that balances operational efficiency with strategic advantage, thereby providing actionable insights for market design and participant strategy.
- 28 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessThe optimal mechanism in high-frequency batch trading with multiple news
We develop a new high-frequency batch model based on game theory to analyze market behavior and address the issue of optimal trading strategies under multiple news arrival scenarios. In this framework, the behaviors and the profits of high-frequency traders (HFTs) are critically influenced by both the quantity and the sequence of news arrival. The model incorporates three interconnected games among HFTs and is refined to identify the trembling-hand perfect Nash equilibrium (THPNE). By maximizing the total market welfare measure, an optimal batch interval is determined to achieve the most efficient THPNE within the batch mechanism. Within this equilibrium, we characterize the optimal spread strategy and dynamic order strategy of HFTs. Single- and multi-period batch models are constructed to compute payoff matrices by analyzing market microstructure and HFTs’ behavior across multiple sequential news arrival scenarios, thereby enabling the derivation of the market spread and HFT profits. Finally, we show the optimal batch interval, a key practical finding that balances operational efficiency with strategic advantage, thereby providing actionable insights for market design and participant strategy.
- 31 March 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 321
PrivTabRAG: A data-model co-design framework for privacy-preserving retrieval-augmented generation on tabular data
- 31 March 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 321
PrivTabRAG: A data-model co-design framework for privacy-preserving retrieval-augmented generation on tabular data
- 2 March 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in INFORMS Journal on Computing
An Adaptive Federated Learning Algorithm with Attenuated Memory on Non-IID and Long-Tail Data
Data sharing and privacy protection in the business sector present dual challenges, especially in the financial and medical fields. We propose an adaptive federated learning algorithm with attenuated memory (AFLAM) to address three critical problems in federated learning: nonindependently and nonidentically distributed (non-IID), long-tail distribution, and privacy. AFLAM dynamically recalculates each client’s weight with attenuated memory of the gradient history to mitigate bias from non-IID data, enhancing the modeling ability of tail data. AFLAM protects privacy by transmitting scaled instead of true gradient information. Two AFLAM algorithms, client-based and parameter-based, are proposed. AFLAM performs better with non-IID and long-tail distribution than existing state-of-the-art methods, improving accuracy by up to 5.71%. Our algorithm provides a novel approach to data sharing and privacy protection challenges. History: Accepted by Ram Ramesh, Area Editor for Data Science & Machine Learning. Funding: H. Yang is grateful for financial support from the National Natural Science Foundation of China [Grant 71771006] and the Beijing Jianlong Heavy Industry Program [Grant 20251202]. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2024.0765 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2024.0765 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .
- 2 March 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in INFORMS Journal on Computing
An Adaptive Federated Learning Algorithm with Attenuated Memory on Non-IID and Long-Tail Data
Data sharing and privacy protection in the business sector present dual challenges, especially in the financial and medical fields. We propose an adaptive federated learning algorithm with attenuated memory (AFLAM) to address three critical problems in federated learning: nonindependently and nonidentically distributed (non-IID), long-tail distribution, and privacy. AFLAM dynamically recalculates each client’s weight with attenuated memory of the gradient history to mitigate bias from non-IID data, enhancing the modeling ability of tail data. AFLAM protects privacy by transmitting scaled instead of true gradient information. Two AFLAM algorithms, client-based and parameter-based, are proposed. AFLAM performs better with non-IID and long-tail distribution than existing state-of-the-art methods, improving accuracy by up to 5.71%. Our algorithm provides a novel approach to data sharing and privacy protection challenges. History: Accepted by Ram Ramesh, Area Editor for Data Science & Machine Learning. Funding: H. Yang is grateful for financial support from the National Natural Science Foundation of China [Grant 71771006] and the Beijing Jianlong Heavy Industry Program [Grant 20251202]. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2024.0765 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2024.0765 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .
- 10 February 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in INFORMS Journal on Computing
Code and Data Repository for An Adaptive Federated Learning Algorithm with Attenuated Memory on Non-IID and Long-tail Data
Addressing the dual challenges of privacy protection and data sharing in sectors such as finance and healthcare, this repository proposes AFLAM (Adaptive Federated Learning Algorithm with Attenuated Memory). By leveraging the decaying mechanism of gradient history to dynamically adjust client weights, this method effectively tackles data heterogeneity and significantly enhances the overall performance of the federated learning model.
- 10 February 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in INFORMS Journal on Computing
Code and Data Repository for An Adaptive Federated Learning Algorithm with Attenuated Memory on Non-IID and Long-tail Data
Addressing the dual challenges of privacy protection and data sharing in sectors such as finance and healthcare, this repository proposes AFLAM (Adaptive Federated Learning Algorithm with Attenuated Memory). By leveraging the decaying mechanism of gradient history to dynamically adjust client weights, this method effectively tackles data heterogeneity and significantly enhances the overall performance of the federated learning model.
- 30 January 2026
- Published by Elsevier in Neurocomputing
- Vol. 674
Quantifying observable privacy in differentially private generative models under black-box access
- 30 January 2026
- Published by Elsevier in Neurocomputing
- Vol. 674
Quantifying observable privacy in differentially private generative models under black-box access
- Open Access
- 22 January 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (2) , 388
Differentially Private Generative Modeling via Discrete Latent Projection
Deep generative models trained on sensitive data pose significant privacy risks, yet enforcing differential privacy (DP) in high-dimensional generators often leads to severe utility degradation. We propose Differentially Private Vector-Quantized Generation (DP-VQG), a three-stage generative framework that introduces a discrete latent bottleneck as the interface for privacy preservation. DP-VQG separates geometric structure learning, differentially private discrete latent projection, and non-private prior modeling, ensuring that privacy-induced randomness operates on a finite codebook aligned with the decoder’s effective support. This design avoids off-support degradation while providing formal end-to-end DP guarantees through composition and post-processing. We provide a theoretical analysis of privacy and utility, including explicit bounds on privacy-induced distortion. Empirically, under the privacy budget of , DP-VQG attains Fréchet Inception Distance (FID) scores of 18.21 on MNIST and 77.09 on Fashion-MNIST, surpassing state-of-the-art differentially private generative models of comparable scale. Moreover, DP-VQG produces visually coherent samples on high-resolution datasets such as Flowers102, Food101, CelebA-HQ, and Cars, demonstrating scalability beyond prior end-to-end DP generative approaches.- 22 January 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (2) , 388
Open AccessDifferentially Private Generative Modeling via Discrete Latent Projection
Deep generative models trained on sensitive data pose significant privacy risks, yet enforcing differential privacy (DP) in high-dimensional generators often leads to severe utility degradation. We propose Differentially Private Vector-Quantized Generation (DP-VQG), a three-stage generative framework that introduces a discrete latent bottleneck as the interface for privacy preservation. DP-VQG separates geometric structure learning, differentially private discrete latent projection, and non-private prior modeling, ensuring that privacy-induced randomness operates on a finite codebook aligned with the decoder’s effective support. This design avoids off-support degradation while providing formal end-to-end DP guarantees through composition and post-processing. We provide a theoretical analysis of privacy and utility, including explicit bounds on privacy-induced distortion. Empirically, under the privacy budget of , DP-VQG attains Fréchet Inception Distance (FID) scores of 18.21 on MNIST and 77.09 on Fashion-MNIST, surpassing state-of-the-art differentially private generative models of comparable scale. Moreover, DP-VQG produces visually coherent samples on high-resolution datasets such as Flowers102, Food101, CelebA-HQ, and Cars, demonstrating scalability beyond prior end-to-end DP generative approaches. - 3 December 2025
- Published by Elsevier in Information Sciences
- Vol. 732
Enhancing privacy-preserving data synthesis utility with two-stage diffusion models
- 3 December 2025
- Published by Elsevier in Information Sciences
- Vol. 732
Enhancing privacy-preserving data synthesis utility with two-stage diffusion models
- Open Access
- 1 December 2025
- Published by KEAI Publishing in Journal of Management Science and Engineering
- Vol. 10 (4) , 623-636
Why do investors pay for additional expenses? A limited learning simulation for the expense ratio dispersion of active mutual funds
- 1 December 2025
- Published by KEAI Publishing in Journal of Management Science and Engineering
- Vol. 10 (4) , 623-636
Open AccessWhy do investors pay for additional expenses? A limited learning simulation for the expense ratio dispersion of active mutual funds