Jelena Bradic is an Assistant Professor of Statistics at the University of California at San Diego associated with the  Mathematics Department. She directs the Statistical Lab for Learning Large-Scale and Complex Data. Her interests are in machine learning, high dimensional statistics, ensemble learning, robust statistics and survival analysis. Her application areas are in big data analysis, time-to-event analysis, econometrics and computational  biology.
Prof. Bradic holds a  PhD degree in Statistics from Princeton University (2011) associated with the Operations Research and Financial Engineering Department and under the guidance of Prof Jianqing Fan. 

​Her undergraduate and masters degree are in Probability and Statistics from Belgrade University's  School of Mathematics, Serbia.   ​
Prof. Bradic has been awarded a 2017 NSF Grant (sole PI), awarded by the NSF Division of Mathematical Sciences, a 5-year partnership with Scientific Research Community of the Research Foundation Flanders (FWO),  she is the recipient of the 2014 Hellman Fellowship,  awarded by the Hellman Foundation, she has been awarded a 2012 NSF Grant award (sole PI), awarded by the NSF Division of Mathematical Sciences, she has been the recipient of the 2012 WCAI Research Grant Award, awarded by the University of Pennsylvania ​and the 2010 LAHA Award, awarded by the Institute of Mathematical Statistics. Prof. Bradic has been recognized as an outstanding teacher  at Princeton University with APGA Award for Teaching and Scholarly Excellence and an Excellence in Teaching Award. She currently serves on committees for National Science Foundation and the American Statistical Association.


Prof. Bradic's projects are centered around:
​
  • Robust Statistics:  analyzing and proposing methods adaptive to the presence of extraneous or adversarial noise, heavy-tailed distributions or sparsity assumptions;

  • Large Scale Statistics: sparse recovery and prediction accuracy for  large scale regression including computational tradeoffs and data heterogeneity;

  • Inference in High-Dimensional Models: developing methods for testing complex statistical hypothesis that are able to do well  with model misspecification or corrupted observations;

  • Ensemble Learning:  methods based on aggregation of noisy or imperfect learners including adversarial and agnostic learning;                        ​
​
  • High-Dimensional Censored Data Models: estimation of the parameters in the models involving time-to-event data with censoring, left-censoring, quantiles, right-censoring.

​