I'm trying to come up with a generative model that can input a name and output all valid formats of it.

For example, "Bob Dylan" could be an input and the gen model will output "Dylan, Bob", "B Dylan", "Bob D" and any other type of valid formatting of a person's name. So given my example the gen model doesn't seem that complicated to build, but it also has to handle stuff like "Dylan, Bob" and "B Dylan", but obviously the 2nd one it shouldn't output "Bob Dylan" as a potential output cause inferring that requires more than just "B Dylan". Any ideas for a good Generative Model for this?

  • $\begingroup$ Do you really need a generative model to do this? $\endgroup$ – nbro Oct 20 '19 at 0:27
  • $\begingroup$ mmm well not if you can think of a better way? $\endgroup$ – Dylan Y Oct 20 '19 at 3:34
  • $\begingroup$ @nbro if your looking some fun why not $\endgroup$ – mshlis Oct 20 '19 at 4:13
  • $\begingroup$ @DylanY I would use an encoder decoder network like in translation systems and I’d induce a prior based on inputs (maybe a secondary step). Seems like a fun problem $\endgroup$ – mshlis Oct 20 '19 at 4:15
  • $\begingroup$ like a VAE? What did you have in mind for doing it? $\endgroup$ – Dylan Y Oct 20 '19 at 4:32

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.