The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks

GPTKB entity

Statements (15)
Predicate Object
gptkbp:instanceOf gptkb:academic_journal
gptkbp:author gptkb:Jonathan_Frankle
gptkb:Michael_Carbin
gptkbp:citation highly cited
gptkbp:contribution proposes the lottery ticket hypothesis
gptkbp:field gptkb:machine_learning
deep learning
gptkbp:foundIn dense, randomly-initialized neural networks contain subnetworks that can be trained in isolation to reach comparable accuracy
https://www.w3.org/2000/01/rdf-schema#label The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks
gptkbp:method iterative pruning
gptkbp:publicationYear 2019
gptkbp:publishedIn gptkb:International_Conference_on_Learning_Representations
gptkbp:url https://arxiv.org/abs/1803.03635
gptkbp:bfsParent gptkb:Lottery_Ticket_Hypothesis
gptkbp:bfsLayer 8