Knowledge base: Warsaw University of Technology

Settings and your account

Back
User profile photo Kamil Deja

Expert menu

Profile

Contacts

  • Room no
    413
  • Communication languages
    Polish

Achievements summary

Bibliometry*

Biogram

Kamil Deja is a 4th-year Ph.D. Student at Warsaw University of Technology. His research focuses on Machine Learning with special interests in Generative Modelling. He is a member of the ALICE Collaboration at CERN, where with support from the NCN Preludium grant, he develops novel machine learning based methods for fast simulation of particle collisions. Kamil also works on novel continual learning methods. He also has industrial experience in international corporations DXC Technology and Amazon, where during a summer research internship in 2021, he worked on state-of-the-art text-to-speech machine learning solutions. His research resulted in several first-authored publications, among the others: BinPlay: A Binary Latent Autoencoder for Generative Replay Continual Learning, IJCNN 2021 (IF: 7.09), End-to-end Sinkhorn Autoencoder with Noise Generator, IEEE Access (IF: 3.37), Multiband-VAE, and multiple presentations including major international venues such as LHCp 2020 Puebla (Mexico), MLSE Pittsburg (USA) and IML in Geneva (Switzerland).

Request ORCID revoke

Uniform Resource Identifier
https://repo.pw.edu.pl/info/author/WUT227dd513dc1847f08373dca9598777f6/
URN
urn:pw-repo:WUT227dd513dc1847f08373dca9598777f6

* presented value of the Hirsch indexOpening in a new tab is approximative calculation obtained in the Repository based on the scientist's publications (including autocitations) in the Repository and Internet information analysis. The value is close to the value obtained with the Publish or Perish system. In general it is higher than the value given by the Scopus or Web of ScienceOpening in a new tab sites. In the case of undervalued number, first of all take care of completeness of the Repository.


Log in

Please log in first.

Confirmation
Are you sure?
Report incorrect data on this page
clipboard