Hi. I'm Iñigo Urteaga.

I am an Associate Research Scientist in the Applied Math department at Columbia University and the Data Science Institute. I am currently working on descriptive, predictive, and prescriptive modeling for electronic health records, in collaboration with Prof. Chris Wiggins and Prof. Noémie Elhadad

I have specialized in statistical data processing, Bayesian Theory, approximate (Monte Carlo and Variational) inference methods, and reinforcement learning (e.g. the multi-armed bandit). My research focuses not only on the development of Bayesian probabilistic models, but also on their application to a wide range of disciplines.

I completed my Ph.D. in Electrical Engineering at Stony Brook University under the supervision of Prof. Petar M. Djurić. My dissertation is entitled "Sequential Monte Carlo methods for inference and prediction of time-series". My research attracted interest in econometrics (for prediction in stochastic volatility models) and biomedicine (for the study of fetal heart-rate signals).

These are my main areas of research.

Bayesian Theory

I am interested in probabilistic modeling in general, and Bayesian statistics in particular. I study Bayesian methods and their application to a plethora of problems.

Monte Carlo Methods

I have specialized in Sequential Monte Carlo methos (i.e., Particle Filters), and I am interested in related techniques such as Adaptive Importance Sampling and advanced MCMC methods.

Stochastic Processes

My dissertation is on the description, estimation and prediction of stochastic processes. I have extensively worked with ARMA, FARIMA and related models for time-series data.

Non-parametric Bayesian models

I am interested in the adoption of non-parametric Bayesian models (e.g., Gaussian, Dirichlet and Pitman-Yor processes) for data clustering and estimation.

The multi-armed bandit problem

I have been recently working on the Bayesian analysis of the multi-armed bandit problem, by exploring the use of approximate inference (IS and VI) for more flexible reward modeling.

Reinforcement learning & prescriptive modeling

I am interested in reinforcement learning in general, and its application to policy evaluation and prescriptive modeling. In particular, in the context of the healthcare process.

Here is a timeline of my professional experience.

  • Associate Research Scientist

    Applied Math department
    Columbia University
  • Postdoctoral Research Scientist

    Columbia University
  • Ph.D. Electrical Engineering

    Stony Brook University
  • Researcher

    Tecnalia Telecom
  • Telecommunication Engineer

  • Research Assistant

    Colorado School of Mines
  • M.S. Telecommunication Engineering

    ETSI Bilbao (UPV/EHU)

I am honored to have received ...

2016 Best Graduate EE Student award.

Best Graduate Student in the Electrical and Computer Engineering department at Stony Brook University

Armstrong Memorial Research Foundation

Spring 2016 Provost Graduate Lecture Series Speaker.

Provost Graduate Lecture talk at Stony Brook University

Lecture available online Youtube

Fall 2015 Distinguished Travel Award.

For outstanding research by Stony Brook University graduate students

Distinguised Travel Award by the Stony Brook Graduate School

2015 Professional Development Award.

GSEU award for developing Graduate Student's full professional potential.

New York State and Graduate Student Employees Union

Torres Quevedo Research Fellowship.

2009-2011 fellowship PTQ-09-02-01814 for young researchers

Ministerio de Ciencia e Innovación, España

Find more details in my full CV

Download my CV

You can contact me via email

inigo.urteaga @ columbia.edu

Or find me on ...