Zoltán Puha

Zoltán Puha

PhD student in Statistics/Data Science

Zalando SE

Biography

Zoltán is a PhD at JADS under the supervision of Prof. Maurits Kapten and dr. Aurelie Lemmens. His main focus is on combining Active Learning and Causal Learning.

Interests

  • Active Learning
  • Bayesian Tree based methods
  • Reinforcement Learning

Education

  • PhD in Statistics, 2018-

    Tilburg University

  • MSc in Economics, 2016

    University of Amsterdam

  • BA in Applied Economics, 2012-2015

    Eötvös Loránd University

Experience

 
 
 
 
 

Data Scientist

Dutch Analytics

Jul 2017 – Dec 2017 Delft, NL

Worked on Deep Learning for predictive maintenance

Responsibilities include:

  • Model building
  • Test case creation
 
 
 
 
 

Data Analyst

Wakoopa

Jun 2016 – Jun 2017 Amsterdam, NL
Analyzing massive amount of behavioral data in the field of marketing research

Accomplish­ments

1st place in EUHackathon

European Parliament Nov 2016

Skills

Machine Learning

After graduating from my Masters, I started to focus on working in ML related fields and after 1.5 years of professional experience I returned to academia to pursue a PhD. My day-to-day research still involves working closely with companies, but I have also TAd for courses, supervised MSc students and taught introductory programming.

Manual-lever espresso

I enjoy making good espresso with a 20 year-old La Pavoni

Recent Posts

Paper accepted to IncrLearn @ ICDM'20

Paper at IncrLearn
Last updated on 2020-09-22 1 min read

Preprint of illegal download paper available

Open Access is not a Panacea, even if it’s Radical – an Empirical Study on the Role of Shadow Libraries in Closing the Inequality of Knowledge Access. - Balazs Bodo, Daniel Antal, Zoltan Puha The preprint of our paper with Balazs Bodo and Daniel Antal is available and have been submitted for peer review.
Last updated on 2020-07-09 1 min read

Talks

InrLearn@ICDM 2020

Active Learning for ITE estimation
2020-11-17 00:00 Virtual
Slides

SCECR 2019

Using the uncertainty of causal inference estimates to plan experiments.
2019-06-08 00:00 Hong Kong
Slides

Contact