Explain 'Pruning' in the context of Neural Networks.

Explain 'Pruning' in the context of Neural Networks.

Approach: Describe removing unnecessary weights or neurons to compress the model.

The process of removing weights or entire neurons from a trained neural network that contribute little to the output. This reduces the model size and improves inference speed with minimal loss in accuracy.

Follow-up: What is the 'Lottery Ticket Hypothesis'?

Key concepts: model compression, inference speed, sparsity.