TA.Talal Alrawajfeh
← All research

Ongoing

Interpretable Differentially Private Continual Learning with Prototype-Based Classifiers

Prototype-based classification for interpretable continual learning under differential privacy. Submitted to ICLR.

Authors
Yuan He, Talal Alrawajfeh, Wachirapatra Mahatubgris, Agil Aghasanli, Mikko A. Heikkilä, Plamen P Angelov, Antti Honkela
Venue
Submitted to ICLR
Status
Ongoing

Learning continually with interpretable classifiers

  1. Evolving tasksLearning from a sequence of datasets
  2. Prototype classifiersRepresent classes through prototypes
  3. Research goalCombine interpretability with differential privacy
This collaboration studies prototype-based classification for interpretable continual learning under differential privacy.

A collaboration on interpretable differential privacy in continual learning, using prototype-based classifiers. The paper has been submitted to ICLR. Public paper and code links will be added when available.