Me

Kevin Bello

Department of Computer Science

Purdue University

Office: Lawson Building 2149 #29

Email: kbellome at purdue dot edu

Resume: PDF (Updated: 09/09)

Summary

I am a PhD student in the Department of Computer Science at Purdue University, working with Prof. Jean Honorio. I am broadly interested in Artificial Intelligence and Machine Learning. My research focuses on developing algorithms that are computationally and statistically efficient for various machine learning problems. Recently, I have worked in structured prediction studying efficient learning with latent variables (NeurIPS'18), minimax bounds (PDF), and conditions for when exact inference is possible in polynomial time (NeurIPS'19).

Previously, I received a BSc in Mechatronics Engineering from the National University of Engineering in Lima, Peru where I was advised by Alberto Coronado.


News

09/2019: Our paper on exact inference was accepted at NeurIPS 2019!

05/2019: Summer internship at Facebook Ads Ranking team, mentored by Yunlong He.

09/2018: Two papers accepted at NeurIPS 2018, the premier conference on machine learning.


Selected Publications

2019:

  1. Minimax Bounds for Structured Prediction. [PDF]
    K. Bello, J. Honorio
    (Under submission.)
  2. Exact Inference in Structured Prediction. [PDF]
    K. Bello, J. Honorio
    NeurIPS'19. (To appear.) In Proceedings of the 33rd Annual Conference on Neural Information Processing Systems, 2019.

2018:

  1. Learning Latent Variable Structured Prediction Models with Gaussian Perturbations. [PDF]
    K. Bello, J. Honorio
    NeurIPS'18. In Proceedings of the 32nd Annual Conference on Neural Information Processing Systems, 2018.
  2. Computationally and Statistically Efficient Learning of Bayes Nets Using Path Queries. [PDF]
    K. Bello, J. Honorio
    NeurIPS'18. In Proceedings of the 32nd Annual Conference on Neural Information Processing Systems, 2018.

Presentations

  • (Upcoming.) Exact Inference in Structured Prediction.
    NeurIPS'19. 33rd Annual Conference on Neural Information Processing Systems, Vancouver, Canada.
  • Learning Latent Variable Structured Prediction Models with Gaussian Perturbations.
    NeurIPS'18. 32nd Annual Conference on Neural Information Processing Systems, Montreal, Canada.
  • Computationally and Statistically Efficient Learning of Bayes Nets Using Path Queries.
    NeurIPS'18. 32nd Annual Conference on Neural Information Processing Systems, Montreal, Canada.
  • Labor Market Demand Analysis for Engineering Majors in Peru Using Shallow Parsing and Topic Modeling
    MLSS'15. In Machine Learning Summer School, Poster Session, Kyoto, Japan.

Academic Service

  • Conferences (reviewer): NeurIPS'19.

Teaching

  • Spring 2017: CS 251, Data Structures and Algorithms. (Taught by Prof. Xavier Tricoche.)
  • Fall 2016: CS 251, Data Structures and Algorithms. (Taught by Prof. Elisha Sacks.)