Structure Learning of Markov Networks and Log-linear Models

Apr 1, 2015

Figure by Jan Strappa

This page is under construction.

This is the area in which I started working as a researcher. I have worked as a research assistant since 2013 and later pursued my doctoral studies in the same area. My thesis work is summarized in the paper Efficient comparison of independence structures of log-linear models.

Markov networks Probabilistic Graphical Models Log-linear models Structure learning Machine Learning academic

Jan Strappa

ML Researcher | Data Scientist

My interests include Artificial Intelligence & Machine Learning (in particular, metaheuristics & evolutionary computation), parallel and distributed computing, probabilistic models, data mining, deep learning, and NLP.

Publications

Comparación eficiente de estructuras de independencia de modelos loglineales

My doctoral thesis.

Jan Strappa

Blankets Joint Posterior score for learning Markov network structures

Journal paper: “Blankets Joint Posterior score for learning Markov network structures”

Federico Schlüter, Jan Strappa, Diego H. Milone, Facundo Bromberg

Blankets Joint Posterior score for learning Markov network structures

Efficient comparison of independence structures of log-linear models

Paper under review: “Efficient comparison of independence structures of log-linear models”

Jan Strappa, Facundo Bromberg

The Grow-Shrink strategy for learning Markov network structures constrained by context-specific independences

Iberamia 2014 conference paper– “The Grow-Shrink strategy for learning Markov network structures constrained by context-specific independences”.

Alejandro Edera, Jan Strappa, Facundo Bromberg