Sensitivity-Informed Provable Pruning of Neural Networks

DOI10.1137/20M1383239OpenAlexW4205646602WikidataQ111519483 ScholiaQ111519483MaRDI QIDQ5037562

Lucas Liebenwein, Igor Gilitschenski, Cenk Baykal, Daniela Rus, Dan Feldman

Publication date: 1 March 2022

Published in: SIAM Journal on Mathematics of Data Science (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1137/20m1383239

zbMATH Keywords

compressionpruninggeneralization

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Related Items (4)

Getting away with more network pruning: from sparsity to geometry and linear regions ⋮ PAC-learning with approximate predictors ⋮ Deep Neural Networks Pruning via the Structured Perspective Regularization ⋮ Unnamed Item

Uses Software

• SqueezeNet

Cites Work

• Unnamed Item
• Unnamed Item
• High-Dimensional Probability
• Improved Approximation Guarantees for Packing and Covering Integer Programs
• Deep double descent: where bigger models and more data hurt*
• Fast Convex Pruning of Deep Neural Networks
• Practical algorithms for a family of waterfilling solutions
• A unified framework for approximating and clustering data

This page was built for publication: Sensitivity-Informed Provable Pruning of Neural Networks