On the compression of neural networks using ℓ0-norm regularization and weight pruning.
Felipe Dennis de Resende OliveiraEduardo Luiz Ortiz BatistaRui SearaPublished in: Neural Networks (2024)
Keyphrases
- neural network
- norm regularization
- convex optimization
- canonical correlation analysis
- regularization methods
- total variation
- matrix completion
- image compression
- loss function
- pattern recognition
- sparse representation
- group lasso
- feature weights
- max margin
- multiple kernel learning
- solution path
- contingency tables
- efficient optimization