Lab Affiliation(s):
CSAIL
Post Doc Sponsor / Advisor:
Tommi Jaakkola
Areas of Expertise:
  • Machine Learning
Date PhD Completed:
August, 2012
Expected End Date of Post Doctoral Position:
August 31, 2015

Jean Honorio

  • Post Doctoral

MIT Unit Affiliation: 

  • Electrical Engineering & Computer Science

Lab Affiliation(s): 

CSAIL

Post Doc Sponsor / Advisor: 

Tommi Jaakkola

Date PhD Completed: 

Aug, 2012

Top 3 Areas of Expertise: 

Machine Learning

Expected End Date of Post Doctoral Position: 

August 31, 2015

CV: 

Research Projects: 

Modern statistical problems are high dimensional (big data). My research in this area focus on developing computationally and statistically efficient algorithms, understanding their behavior using concepts such as convergence, sample complexity, and privacy, and designing new modeling paradigms such as models rooted in game theory. My theoretical and algorithmic work is directly motivated by, and contributes to, applications in political science (affiliation and influence), neuroscience (brain disorders such as addiction), and genetics (diseases such as cancer).

Thesis Title: 

Tractable Learning of Graphical Model Structures from Data

Thesis Abstract: 

This thesis focuses on tractable learning of probabilistic as well as game-theoretical graphical model structures from data. Contributions include convergence analysis of biased stochastic optimization, generalization guarantees, proposing domain-specific priors and developing efficient optimization algorithms.

Top 5 Awards and honors (name of award, date received): 

Best Paper Award: "Predicting Cross-task Behavioral Variables from fMRI Data Using the k-Support Norm". September 27, 2014, Medical Image Computing and Computer-Assisted Intervention (MICCAI), Workshop on Sparsity Techniques in Medical Imaging, Boston.
Top Graduate Student Award 2012. February 2013, Stony Brook University, New York.
Provost's Graduate Student Lecture Series: "Learning Gaussian Graphical Models with Domain Specific Priors". May 2, 2012, Stony Brook University, New York.
Catacosinos Computer Science Award (US$ 7,500). Fall 2011. Stony Brook University, New York.
Institute for Biomedical Engineering Award (US$ 8,300). Spring 2005, George Washington University, Washington DC.

5 Recent Papers: 

"Learning the Structure and Parameters of Large-Population Graphical Games from Behavioral Data".
Honorio J., Ortiz L.
Journal of Machine Learning Research (JMLR), 2014. (accepted, pending publication)

"A Unified Framework for Consistency of Regularized Loss Minimizers". (presentation)
Honorio J., Jaakkola T.
International Conference on Machine Learning (ICML). Beijing/China, 2014. (Acceptance rate: 25%)

"Tight Bounds for the Expected Risk of Linear Classifiers and PAC-Bayes Finite-Sample Guarantees".
Honorio J., Jaakkola T.
Artificial Intelligence and Statistics (AISTATS). Reykjavik/Iceland, 2014. (Acceptance rate: 35.8%)

"Inverse Covariance Estimation for High-Dimensional Data in Linear Time and Space: Spectral Methods for Riccati and Sparse Models".
Honorio J., Jaakkola T.
Uncertainty in Artificial Intelligence (UAI). Washington, 2013. (Acceptance rate: 31.3%)

"Two-Sided Exponential Concentration Bounds for Bayes Error Rate and Shannon Entropy". (presentation)
Honorio J., Jaakkola T.
International Conference on Machine Learning (ICML). Atlanta, 2013. (Acceptance rate: 23.5%)

Contact Information: