I’m Yanzhi Wang (王衍之), a third year PHD student at the College of Engineering, Peking University, fortunate to be advised by Prof. Jie Song(Google Scholar) and Prof. Jianxiao Wang(Google Scholar). Prior to this, I received my Bachelor’s degrees in Science and Economics from Peking University. My research interests include data centric AI, data-driven optimization, energy systems and uncertainty estimation. You can find my Google Scholar profile here.

Recently, my research centers on advancing the interpretability of data science within energy systems. I am particularly interested in leveraging AI and data-driven methodologies to implement end-to-end optimization. My current work emphasizes enhancing the reliability and efficiency of energy systems under uncertainty through data-centric decision-making.

You can find my CV here: [PDF]

Publications (Article)

• [Engineering] Data-Centric Optimization: An End-to-End Strategy for Power Systems

  Yanzhi Wang, Jianxiao Wang*, Jie Song* (2025)

  [PDF]

• Proposed DCOpt, a framework valuing features by their economic impact in end-to-end optimization.
• Developed a rolling-horizon algorithm that reduces decision costs by 8.9% in OOD scenarios.
• Demonstrated that curating high-value features outperforms increasing data quantity using ISO-NE data.

• [IEEE Transactions on Industry Applications] Feature Selection for Battery Lifetime Prediction using Explainable Machine Learning

  Yanzhi Wang, Jianxiao Wang*, Xi Chen, Yishen Wang, Jie Song (2025)

  [PDF]

• [Nexus] Bridging Prediction and Decision: Advances and Challenges in Data-Driven Optimization

  Yanzhi Wang, Jianxiao Wang*, Haoran Zhang, Jie Song* (2025)

  [PDF]

• Analyzed frameworks for integrating prediction and decision-making.
• Compared advancements in multidimensional data-driven optimizations
• Addressed challenges in data input, modeling, and decision processes
• Highlighted applications in power systems, management, and control

• [IEEE Transactions on Industrial Informatics] Data Purification for Improved Power Dispatch Against Renewable Uncertainty

  Yanzhi Wang, Jianxiao Wang, Jie Song* (2025)

  [PDF]

• [Patterns]Unveiling Value Patterns via Deep Reinforcement Learning in Heterogeneous Data Analytics (Cover of the May 2024 Issue)

Yanzhi Wang, Jianxiao Wang*, Feng Gao, Jie Song* (2024)

[PDF]

• We introduce a learning-based paradigm for data valuation across scenarios.
• Our study identifies varying effects of high/low-quality data on model efficacy.
• This method explores inherent and transferable value patterns across datasets.
• Analysis reveals data value’s geographic sensitivity in nationwide power forecasts.

• [IEEE Transactions on Industry Applications] Dissecting Renewable Uncertainty via Deconstructive Analysis-Based Data Valuation

  Yanzhi Wang, Jie Song* (2024)

  [PDF]

• [Management Review (In Chinese)] Research on the Circulation and Revenue Sharing Mechanisms of Data Elements: An Example of Integrating Meteorological Data in Wind Power Scenarios, Yanzhi Wang, Jingsi Huang, Jianxiao Wang, Feng Gao, Jie Song* (2024)

  [PDF]

• [Chinese Journal of Management Science (In Chinese)] Research on Data Value Based on Demand Forecast of Online Medical Platform, Zhao Yue, Yanzhi Wang, Jingsi Huang, Jie Song*, Xuan He (2024)

  [PDF]

Publications (Conference)

• [IAS 2024] Feature Valuation Toward Improved State Estimation for Automotive Lithium-ion Battery

  Yanzhi Wang, Jianxiao Wang*, Jie Song

  [PDF]

• [CASE 2022] What Drives Patients to Choose a Physician Online? A Study based on Tree Models and SHAP Values

  Yanzhi Wang, Yue Zhao, Jie Song*, Hongju Liu

  [PDF]

Manuscripts under submission

• AI-Powered Data Valuation for Robust Financial Modelling

  Yanzhi Wang, Jianxiao Wang, Yaqin Hu, Jie Song, Xiaofei Zhao. (In Submission)

• Condensed Data Acquisition for Non-Intrusive Load Monitoring via Household Selection

  YiHang Jin, Yanzhi Wang, Jianxiao Wang*, et al. (Under Review in CSEE Jornal of Power and Energy Systems)

Honors and Awards

• 2025.09 National Scholarship
• 2025.05 Presidential Scholarship (Peking University)
• 2024.10 The Youth Talent Support Program of the China Association for Science and Technology (Doctoral Program)
• 2024.09 National Scholarship
• 2024.06 Wei-Ming PhD Scholars (One of 13 recipients selected university-wide)
• 2024.05 Academic Rising Star (One of 5 recipients selected college-wide)
• 2022.01 Outstanding Undergraduate Research Award (College level)

Skills

• Programming: Python, MATLAB, R.
• Academic Training: Advanced Theory of Probability, Advanced Operations Research, Optimization Methods in Machine Learning, Reinforcement Learning, Distributed Optimization, Design and Analysis of Algorithms, Carbon Neutrality and Energy Internet, etc.