Timezone: »

Movement Pruning: Adaptive Sparsity by Fine-Tuning
Victor Sanh · Thomas Wolf · Alexander Rush

Tue Dec 08 09:00 AM -- 11:00 AM (PST) @ Poster Session 1 #288

Magnitude pruning is a widely used strategy for reducing model size in pure supervised learning; however, it is less effective in the transfer learning regime that has become standard for state-of-the-art natural language processing applications. We propose the use of movement pruning, a simple, deterministic first-order weight pruning method that is more adaptive to pretrained model fine-tuning. We give mathematical foundations to the method and compare it to existing zeroth- and first-order pruning methods. Experiments show that when pruning large pretrained language models, movement pruning shows significant improvements in high-sparsity regimes. When combined with distillation, the approach achieves minimal accuracy loss with down to only 3% of the model parameters.

Author Information

Victor Sanh (Hugging Face 🤗)
Thomas Wolf (🤗 Hugging Face)
Alexander Rush (Cornell University)

More from the Same Authors