HIGHLIGHTS
- who: Ryan Bluteau and Robin Gras from the Computer Science, University of Windsor, Sunset Ave, Windsor, ON N B , Canada have published the paper: Lottery Ticket Search on Untrained Models with Applied Lottery Sample Selection, in the Journal: (JOURNAL)
- what: The authors show that the authors can remove up to 95% of the training dataset to discover tickets while still maintaining similar accuracy. The authors propose a hypothesis similar to the ticket hypotheses where a subsample of samples of the training set can train a model with equivalent performance to the original dataset. The authors . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.