Given a set of time series data that are generated from different sites where all sites are investigating the same objective but with slightly different protocols.

Is it possible to use adversarial learning to learn site invariant features for a classification problem, that is, how can adversarial learning be used to minimize experimental differences (e.g. different measurement equipment) so that the learned feature representations from the time series are homogenous for a classification problem?

I have come across multi-domain adversarial learning, but I'm not sure if this is the best formulation for my problem.

  • $\begingroup$ Hi and welcome to this community! What do you mean by "sites" in this context? $\endgroup$ – nbro Jul 6 '19 at 12:53
  • 1
    $\begingroup$ HI, thank you. Sites in this case means different locations around the world. In a given location L_i for example, there are N_i subjects (i=1,..) and each subject has numeric information measured. The numeric information is the same but the way the numeric information is measured is slightly different. In one location, the data for a given subject may be measured for say 2 minutes, and in another it is 1 minute. $\endgroup$ – Mwiza Kunda Jul 6 '19 at 16:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.