AP: Selective Activation for De-sparsifying Pruned Networks

The rectified linear unit (ReLU) is a highly successful activation function in neural networks as it allows networks to easily obtain sparse representations, which reduces overfitting in...

Connectivity Matters: Neural Network Pruning Through the Lens of Effective Sparsity

Max-Affine Spline Insights Into Deep Network Pruning

State-of-the-art (SOTA) approaches to deep network (DN) training overparametrize the model and then prune a posteriori to obtain a ``winning ticket'' subnetwork that can achieve high accuracy....