Kun H.

I'm a Ph.D. candidate in Machine Learning and Data Science with a passion for creating intelligent and efficient solutions. I've had the opportunity to apply my skills in a variety of contexts, including developing an AI-powered system for engineering drawings that achieving an 90% increase in metadata extraction efficiency. and building secure web platforms with AWS S3 integration. I also have experience in DevOps and Cloud Automation, including writing Python scripts for asset verification, administering AWS services, and managing Ubuntu servers. My research focuses on trustworthy AI, specifically robust and interpretable time series analysis, and I have a strong background in deep learning, including transformer, graph neural networks and attention mechanisms. I am committed to developing robust and reliable solutions for real industrial or real life. Core Strengths * Time-Series & Sequential Modelling: Transformers, GNNs, MIL, attention mechanisms, noisy-label learning