CV

PhD in Computer Science

University of Birmingham (UoB)

School of Computer Science

• Research interests: Machine Learning and Optimization, including deep learning, meta learning, Bayesian optimization, reinforcement learning, evolutionary computation, statistical learning, time series analysis, stochastic processes, and their applications in real-world problems.
• PhD thesis: Surrogate-Assisted Evolutionary Algorithms for Computationally Expensive Optimization Problems.
• Primary supervisor: Prof. Xin Yao (relocated to a position outside the UK in 2017).
• Secondary supervisor: Prof. Joshua Knowles (left UoB due to illness in 2017).
• Completed PhD on my own after both supervisors departed UoB, demonstrating ability to conduct independent research.

BSc in Computer Science

南京大学 (Nanjing University, NJU)

Department of Computer Science & Technology

• Selected courses: Calculus, linear algebra, probability & statistics, programming, data structures, operating systems, algorithm, database, data mining, pattern recognition, mathematical modeling, digital image processing, computer network.
• Supervisor: Prof. Yang Yu.

Research Fellow

Department of Economics, University of Warwick

• Line managers: Dr. Amrita Kulka (assistant professor) and Dr. Nikhil Datta (assistant professor).
• Situation: The research project requires scraping large scale complex data from 430+ idiosyncratic online platforms.
• Task: Develop, deploy, and maintain hundreds of scrapers on AWS, process data for academic analysis.
• Action and Result: - Built and deployed 200+ production scrapers across 10+ frameworks with Python (scrapy, selenium, pandas), extracting data and documents from over 12M planning applications. - Developed pattern recognition (tensorflow, keras, scikit-learn) and network solutions for diverse reCAPTCHA puzzles. - Employed ML and CV techniques (e.g. OCR) to analyse texts from PDF documents with 99%+ accuracy.

Machine Learning Researcher

Ford Motor Company

• Line manager: Dr. Yan Wang (technical expert).
• Situation: Engine calibration is a time-consuming process of optimizing engine settings to achieve optimal performance.
• Task: Develop efficient modeling methods and model-based optimisation algorithms to accelerate calibration R&D cycles.
• Action: - Used statistical methods (e.g. PCA) to process and analyse large real-world motor engine datasets with Python (numpy, pandas, scikit-learn) to uncover intrinsic patterns between engine settings and engine performance. - Developed ordinal regression, deep kernel learning, and pre-trained models (deep learning, meta-learning, multivariate Gaussian Process) with Python (numpy, tensorflow, pytorch) and Matlab to approximate motor engine performance. - Designed Bayesian optimisation, bilevel optimisation, and few-shot optimisation frameworks to explore feasible solutions with optimal statistical performance (e.g. expected improvement, probability of improvement) on approximation models. - Worked on multi-disciplinary teams with diverse backgrounds, reported to Ford engineers and scientists.
• Result: Achieved a 40% improvement in calibration efficiency and an 80% increase in the number of feasible engine designs.

Research Assistant

Learning And Mining from DatA (LAMDA) Group, Nanjing University

• Supervisor: Prof. Yang Yu.
• Research topics: Machine Learning, Statistical Learning, Bayesian Optimization.
• Action: - Developed a co-training semi-supervised regression approach to actively learn predictive models for problems. - Designed a top-querying strategy integrating these models with CMA optimisation.
• Result: Improved optimisation efficiency by up to 64% across Bayesian optimisation problems.

FSEO: Few-Shot Evolutionary Optimization via Meta-Learning for Expensive Multi-Objective Optimization

Advances in Neural Information Processing Systems 39 (NeurIPS'25)

Engine Calibration With Surrogate-Assisted Bilevel Evolutionary Algorithm

IEEE Transactions on Cybernetics

Internal Combustion Engine Calibration Using Optimization Algorithms

Applied Energy

Domination-Based Ordinal Regression for Expensive Multi-Objective Optimization

Proceedings of the 2019 IEEE Symposium Series on Computational Intelligence (SSCI'19)

Peer Reviewer

IEEE {TEVC, TCYB, TETCI}, Applied Energy, Evolving Systems, etc.

Program Committee Member

ICLR, AAAI

CAGE Research Grant

The ESRC Centre for Advantage in the Global Economy, University of Warwick

NeurIPS Scholar Award

Neural Information Processing Foundation

IEEE CIS Travel Grant

IEEE Computational Intelligence Society

Ford Motor Company (USA) PhD Scholarship (fully-funded)

University of Birmingham

Outstanding Student, Nanjing University

Nanjing University

Top 3%

Outstanding Student, Department of Computer Science & Technology

Nanjing University

First Prize in China National Mathematics Olympiad

Chinese Mathematical Society

Top 3 in Jiujiang City