Brian Trippe, Ph.D.

Doctoral Thesis Title: Bayesian Linear Modeling in High Dimensions: Advances in Hierarchical Modeling, Inference, and Evaluation

Publications while at MIT:

Diffusion probabilistic modeling of protein backbones in 3D for the motif-scaffolding problem

BL Trippe, J Yim, D Tischer, T Broderick, D Baker, R Barzilay, T Jaakkola

arXiv preprint arXiv:2206.04119

Many processors, little time: MCMC for partitions via optimal transport couplings

Tin D. Nguyen, Brian L. Trippe, Tamara Broderick  https://arxiv.org/abs/2202.11258  22 Feb 2022 

For high-dimensional hierarchical models, consider exchangeability of effects across covariates instead of across datasets

B Trippe, H Finucane, T Broderick

https://arxiv.org/abs/2107.06428

Optimal transport couplings of Gibbs samplers on partitions for unbiased estimation

BL Trippe, TD Nguyen, T Broderick

arXiv preprint arXiv:2104.04514 

Confidently Comparing Estimators with the c-value

BL Trippe, SK Deshpande, T Broderick

arXiv preprint arXiv:2102.09705. 

LR-GLM: High-Dimensional Bayesian Inference Using Low-Rank Data Approximations  

 Brian L. Trippe, Jonathan H. Huggins, Raj Agrawal, Tamara Broderick https://arxiv.org/abs/1905.07499 17 may 2019. 

The kernel interaction trick: Fast bayesian discovery of pairwise interactions in high dimensions

R Agrawal, B Trippe, J Huggins, T Broderick

International Conference on Machine Learning, 141-150