Igor Kulev

Igor Kulev

Ph.D. student
Mailing Address:
Room 236, Building INR, I&C department, EPFL
Station 14, CH – 1015 Lausanne, Switzerland
Tel Direct: +41 21 69 36682
Email: igor dot kulev at epfl dot ch

About my research

My research is devoted to discovering how the intervention affects the behavior of an individual (or a process). The aim is to predict how the intervention will affect a new individual with certain characteristics. In difference with more traditional methods for discovering the heterogeneous treatment effect, I focus on more fine-grained representation of the behavior or the individual (described by sequential observations). Having this level of granularity would allow the analysts to understand better the treatment effect and to predict more accurately the individual’s response to the intervention. In my research I work with probabilistic graphical models and deep learning techniques.

Research Interests

Machine Learning, Probabilistic Graphical Models, Deep Learning, Time Series Analysis, Artificial Intelligence, Human Computer Interaction

Research Projects

Education

  • Ongoing: Ph.D. in Computer Science. School of Computer and Communication Sciences, Swiss Federal Institute of Technology in Lausanne (EPFL). Starting from September 2014.
  • M.S. in Electrical Engineering and Information Technologies. Department of Computer Science and Engineering at Ss. Cyril and Methodius University in Skopje, Macedonia. 2011 – 2013
    Specialization: Intelligent Information Systems.
  • B.S. in Electrical Engineering and Information Technologies. Department of Computer Science and Engineering at Ss. Cyril and Methodius University in Skopje, Macedonia. 2007 – 2011

Publications

  • Igor Kulev, Pearl Pu and Boi Faltings. ”Discovering Persuasion Profiles Using Time Series Data.” In NIPS Time Series Workshop 2016. Barcelona, Spain, 2016 [pdf]