معرفی
Ruqi Zhang is an Assistant Professor in the Department of Computer Science at Purdue University. Previously, she was a postdoctoral fellow at the Institute for Foundations of Machine Learning, UT Austin (2021-2022), and earned her PhD in Statistics from Cornell University (2016-2021). Her research focuses on scalable probabilistic methods for machine learning, including alignment of foundation models, uncertainty quantification, and Bayesian deep learning. She has received awards such as the Ross-Lynn Research Scholar Fund and ICML Best Reviewer recognition.
Education:
- PhD in Statistics, Cornell University (2021)
- MS in Computer Science, Cornell University (2021)
- Bachelor of Science in Mathematics, Renmin University of China (2012-2016)
Research Interests:
- Trustworthy AI: Safety of LLMs/VLMs, alignment mechanisms
- Probabilistic Inference: Bayesian methods, MCMC, variational inference
- Generative Models: Diffusion models, energy-based models
- Uncertainty Estimation: Calibration, out-of-distribution detection
Notable Contributions:
- Developed the Discrete Langevin Sampler for high-dimensional discrete spaces
- Pioneered Low-Precision SGLD for efficient Bayesian neural networks
- Designed DP-Fast MH for privacy-preserving Bayesian inference
Teaching:
- CS57800 - Statistical Machine Learning (2022-Present)
- CS37300 - Data Mining and Machine Learning (2024)
- CS59200 - Probabilistic Machine Learning (2022)
۰مقاله ثبتشده
