Chang Ye

Financial Engineering & Machine Learning Researcher
MSc Financial Engineering (Expected 2026) | PhD in Electrical Engineering

📍 Hong Kong
📧 chang.ye@my.cityu.edu.hk
🔗 LinkedIn | GitHub

About Me

I am a researcher and engineer with a strong background in signal processing, machine learning, and graph-based AI, currently transitioning into quantitative finance and fintech. My research focuses on time series analysis, generative AI, and graph neural networks, with applications in financial modeling, source localization, and recommendation systems.

I hold a PhD in Electrical Engineering from the University of Rochester and am currently pursuing a Master of Science in Financial Engineering at the City University of Hong Kong. I have hands-on experience in building automated trading systems, developing AI-driven forecasting models, and improving real-world machine learning systems during my internship at Amazon.

🔧 Technical Skills

• Programming: Python (Pandas, Scikit-learn, PyTorch), MATLAB, SQL
• ML & AI: Graph Neural Networks, Deep Learning, Time Series Forecasting, Generative AI
• Cloud & Tools: AWS (SageMaker, EC2, S3), Google Colab, Jupyter, Git
• Languages: English, Chinese, Cantonese

🚀 Projects

Stock Back-Testing Framework with Generative AI

• Built a Python-based back-testing engine for quantitative trading strategies
• Integrated real-time market data via Yahoo Finance API
• Implemented technical indicators (MACD, RSI, Bollinger Bands) and AI forecasting models like Kronos
• Analyzed performance using Sharpe ratio, Sortino ratio, and benchmark comparisons

Blind Deconvolution on Graphs Net (BDoG-Net)

• Developed a model-based deep learning solution for source localization over networks
• Outperformed state-of-the-art generative models by 7.5% in AUC on real-world datasets

• Paper

  GitHub

📄 Selected Publications

• Ye, C., & Mateos, G. (2025). BDoG-Net: Algorithm Unrolling for Blind Deconvolution on Graphs. IEEE Transactions on Signal and Information Processing over Networks.
• Ye, C., & Mateos, G. (2025). Blind deconvolution of graph signals: Robustness to graph perturbations. IEEE Signal Processing Letters.
• Ye, C., et al. (2024). A tensor decomposition reveals ageing-induced differences in muscle and grip-load force couplings during object lifting. Scientific Reports.

👨‍💻 Experience

Applied Scientist Intern – Amazon

• Improved CTR prediction models for payment product recommendations
• Used logistic regression, random forest, and LightGBM to achieve a 10% AUC improvement
• Built data pipelines with MySQL and Python on AWS

🎓 Education

• MSc Financial Engineering – City University of Hong Kong (2025–2026)
• PhD Electrical Engineering – University of Rochester (2017–2025)
• MSc Electrical Engineering – University of Rochester (2015–2016)
• BSc Physics – Sun Yat-Sen University (2009–2014)

📜 Certifications

• Google Advanced Data Analytics (Coursera)
• Finance & Quantitative Modeling for Analysts – Wharton (Coursera)

Let’s connect! Feel free to reach out via email or LinkedIn.