Topology-Aware In-Run Data Shapley for Decentralized Learning
Research on efficient data valuation in decentralized learning settings with topology-aware modeling.
MSc Student / AI / NLP / Full-stack Development
I am a Master of Science student at The Hong Kong University of Science and Technology, with a Bachelor of Engineering from Tsinghua University, School of Software. My work spans NLP, speech recognition, AI-powered applications, data analysis, and full-stack development.
I enjoy turning research problems into usable systems. My recent work includes data Shapley methods, protein language models, dialect-to-SQL conversion, digital rural management systems, and AI-assisted applications.
Research on efficient data valuation in decentralized learning settings with topology-aware modeling.
Exploring data valuation methods under adaptive optimization dynamics.
Research on improving controllability and generation quality in protein language models.
Conducted research on Chinese dialect-to-SQL methods and authored a bachelor's thesis.
Designed and implemented a digital management system for community showcase, data storage, filtering, and rural voice broadcasting.
Implemented short-distance text transmission between devices via sound waves with custom encoding and decoding mechanisms.
Built a WeChat mini-program for pronunciation scoring, dictionary search, and user analytics; won awards in software innovation competitions.
Full-stack development for TikTok Shop SEO-related systems.
Developed UE4 plugins for artistic text rendering, ray tracing, and PSD file import.
Feel free to reach out for research collaboration, software engineering projects, or AI application development.