Transfer Learning for Structured Pruning under Limited Task Data

Anonymous

TL;DR: We propose a transfer learning based approach to structured pruning when task data is limited. We empirically explore various design choices and make recommendations to practitioners about how to achieve accurate, compact models under limited data

Abstract: Large, pre-trained models are problematic to use in resource constrained applications. Fortunately, task-aware structured pruning methods offer a solution. These approaches reduce model size by dropping structural units like layers and attention heads in a manner that takes into account the end-task. However, these pruning algorithms require more task-specific data than is typically available. We propose a framework which combines structured pruning with transfer learning to reduce the need for task-specific data. Our empirical results answer questions such as: How should the two tasks be coupled? What parameters should be transferred? And, when during training should transfer learning be introduced? Leveraging these insights, we demonstrate that our framework results in pruned models with improved generalization over strong baselines.

Paper Type: long

Research Area: Efficient/Low-Resource Methods for NLP

Contribution Types: Approaches to low-resource settings, Approaches low compute settings-efficiency

Languages Studied: English