A flow-based stopping criterion (InCoP) reduces the training epochs needed during iterative magnitude pruning while preserving final accuracy, compared with LTH and SAP.
CoRR (2023),http://arxiv.org/abs/2302
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Information Consistent Pruning: How to Efficiently Search for Sparse Networks?
A flow-based stopping criterion (InCoP) reduces the training epochs needed during iterative magnitude pruning while preserving final accuracy, compared with LTH and SAP.