305 Academic Research Building
265 South 37th Street
Philadelphia, PA 19104
Research Interests: Statistics and machine learning
Our research interests include problems at the interface of statistics, machine learning, and AI, such as uncertainty quantification, AI safety, robustness, high-dimensional asymptotic statistics, etc.
The group is always looking to expand. We are recruiting PhD students at Penn to work on problems in statistics and machine learning. PhD applicants interested to work with me should mention this on their application. Please apply through the departments of Statistics & Data Science, Computer and Information Science, and the AMCS program, as it gives higher chances for admission.
Education:
Recent news:
Miscellanea:
Talk slides: GitHub. Google Scholar.
Edgar Dobriban and Zhanran Lin (Working), Joint Coverage Regions: Simultaneous Confidence and Prediction Sets.
Behrad Moniri, Seyed Hamed Hassani, Edgar Dobriban (Working), Evaluating the Performance of Large Language Models via Debates.
Yan Sun, Pratik Chaudhari, Ian Barnett, Edgar Dobriban (Working), A Confidence Interval for the ℓ2 Expected Calibration Error.
Yonghoon Lee, Eric Tchetgen Tchetgen, Edgar Dobriban (Working), Batch Predictive Inference.
Sunay Joshi, Shayan Kiyani, George Pappas, Edgar Dobriban, Seyed Hamed Hassani (Working), Likelihood-Ratio Regularized Quantile Regression: Adapting Conformal Prediction to High-Dimensional Covariate Shifts.
Meshi Bashari, Roy Maor Lotan, Yonghoon Lee, Edgar Dobriban, Yaniv Romano (Working), Synthetic-Powered Predictive Inference.
Yonghoon Lee, Edgar Dobriban, Eric Tchetgen Tchetgen (Working), Finding Distributions that Differ, with False Discovery Rate Control.
Georgy Noarov, Soham Mallick, Tao Wang, Sunay Joshi, Yan Sun, Yangxinyu Xie, Mengxin Yu, Edgar Dobriban (Working), Foundations of Top-k Decoding For Language Models.
Patrick Chao, Edgar Dobriban, Seyed Hamed Hassani (Working), Watermarking Language Models with Error Correcting Codes.
Xinmeng Huang, Shuo Li, Edgar Dobriban, Osbert Bastani, Seyed Hamed Hassani, Dongsheng Ding (Working), One-Shot Safety Alignment for Large Language Models via Optimal Dualization.
“For deep, fundamental, and wide-ranging contributions to mathematical statistics and statistical machine learning, including high-dimensional asymptotics (ridge regression, PCA), multiple testing, randomization tests, scalable statistical learning via random projections and distributed learning, uncertainty quantification for machine learning (calibration, prediction sets), robustness, fairness, and Covid-19 pooled testing via hypergraph factorization.”