2 papers
cs.CV2023
A Generalization of Continuous Relaxation in Structured Pruning
Brad Larson, Bishal Upadhyaya, Luke McDermott +1
Deep learning harnesses massive parallel floating-point processing to train and evaluate large neural networks. Trends indicate that deeper and larger neural networks with an incre…
cs.LG2023
Distilled Pruning: Using Synthetic Data to Win the Lottery
Luke McDermott, Daniel Cummings
This work introduces a novel approach to pruning deep learning models by using distilled data. Unlike conventional strategies which primarily focus on architectural or algorithmic…