Generalization in Lottery Tickets
Tim and Sayak discuss the concept of generalization in lottery tickets, questioning whether these subnetworks only excel in specific tasks or have broader applications. They delve into the significance of parameters in pruning networks for transfer learning and the need for further research in formalizing pruning criteria for optimal generalization in downstream tasks.In this clip
From this podcast
Machine Learning Street Talk (MLST)
Sayak Paul
Related Questions