Tin Nguyen

Tin Nguyen

Quantitative Researhcer

Quantbot Technologies LP

Tin Nguyen is a Quantitative Researcher at Quantbot Technologies, LP. Previously, he was a PhD student at MIT EECS, advised by Tamara Broderick. His research developed computational methods for inference and sensitivity detection in Bayesian models.

Topics

Bayesian Statistics Conditional Random Fields Cross-Validation Decision Trees Evaluation Influence Function Markov Chain Monte Carlo Statistical Learning Theory Supervised Learning Unsupervised Learning

Publications

Tin Nguyen, Ryan Giordano, Rachael Meager, Tamara Broderick (2024). Sensitivity of MCMC-based analyses to small-data removal.

PDF

Jenny Y. Huang, David R. Burt, Tin Nguyen, Yunyi Shen, Tamara Broderick (2024). Approximations to worst-case data dropping: unmasking failure modes.

PDF

Tin Nguyen, Ryan Giordano, Rachael Meager, Tamara Broderick (2024). Using gradients to check sensitivity of MCMC-based analyses to removing data. In DiffAE 2024.

PDF

Sameer K. Deshpande, Soumya Ghosh, Tin Nguyen, Tamara Broderick (2024). Are you using test log-likelihood correctly?. In TMLR.

PDF Poster Video

Kaifu Wang, Hangfeng He, Tin Nguyen, Piyush Kumar, Dan Roth (2023). On regularization and inference with label constraints. In ICML 2023.

See all publications

Talks

Thesis defense
MIT EECS PhD thesis defense
May 7, 2024 8:00 AM — 9:00 AM Virtual
Slides Video
Sensitivity of MCMC analyses to data removal
Contributed talk at BAYSM:O 2023
Nov 13, 2023 1:00 PM — Nov 17, 2023 3:00 PM Virtual
Slides
Are you using test log-likelihood correctly?
Poster presentation at ICBINB Workshop at NeurIPS 2022
Dec 3, 2022 8:00 AM — 5:00 PM New Orleans, LA
PDF
Many processors, little time: MCMC for partitions via optimal transport couplings
Invited talk at ISBA 2022
Jun 28, 2022 1:30 PM — 3:00 PM Montreal
Slides
Many processors, little time: MCMC for partitions via optimal transport couplings
Oral presentation at AISTATS 2022
Mar 29, 2022 8:00 AM — 9:00 AM Virtual
PDF Slides
See all events

Projects

*
Parcel Price Prediction
Build supervised learning model and evaluate accuracy for predicting Boston property prices.
Code
Parcel Price Prediction

Skills

R

3 years experience

Python

4 years experience

SQL

1 year experience

Experience

 
 
 
 
 
Quantitative Researcher
Quantbot Technologies LP
June 2024 – Present NY, New York
 
 
 
 
 
Intern
Quantbot Technologies LP
May 2023 – August 2023 NY, New York
 
 
 
 
 
Research Intern
IBM Research
June 2022 – August 2022 Cambridge, MA
 
 
 
 
 
Research Intern
IBM Research
June 2019 – August 2019 Cambridge, MA

Awards

Student Poster Award
New England Statistical Society May 2022
Outstanding Reviewer Award
Neural Information Processing Systems Oct 2021
Student Travel Award
Neural Information Processing Systems Oct 2018

Professional Services

NeurIPS Reviewer

2021, 2022

AISTATS Reviewer

2022

TMLR Reviewer

2022, 2023