Publication

Pruning weightless neural networks

View document

Bibliographic Details
Summary:Weightless neural networks (WNNs) are a type of machine learning model which perform prediction using lookup tables (LUTs) instead of arithmetic operations. Recent advancements in WNNs have reduced model sizes and improved accuracies, reducing the gap in accuracy with deep neural networks (DNNs). Modern DNNs leverage “pruning” techniques to reduce model size, but this has not previously been explored for WNNs. We propose a WNN pruning strategy based on identifying and culling the LUTs which contribute least to overall model accuracy. We demonstrate an average 40% reduction in model size with at most 1% reduction in accuracy.
Main Authors:Susskind, Z.
Other Authors:Bacellar, A. T. L.; Arora, A.; Villon, L. A. Q.; Mendanha, R.; Araújo, L. S. de.; Dutra, D. L. C.; Lima, P. M. V.; França, F. M. G.; Miranda, I. D. S.; Breternitz Jr., M.; John, L. K.
Year:2022
Country:Portugal
Document type:conference output
Access type:open access
Associated institution:ISCTE
Language:English
Origin:Repositório ISCTE

Similar Items

Similar Items