Multirate Training of Neural Networks

TL;DR

This paper introduces a multirate training method for neural networks that updates different parts at varying speeds, significantly reducing training time in transfer learning without sacrificing performance.

Contribution

It presents a novel multirate training scheme for neural networks, enabling faster transfer learning and offering insights into parameter partitioning and convergence analysis.

Findings

Achieves nearly 50% reduction in fine-tuning time

Maintains comparable generalization performance

Provides convergence analysis and parameter splitting strategies

Abstract

We propose multirate training of neural networks: partitioning neural network parameters into "fast" and "slow" parts which are trained on different time scales, where slow parts are updated less frequently. By choosing appropriate partitionings we can obtain substantial computational speed-up for transfer learning tasks. We show for applications in vision and NLP that we can fine-tune deep neural networks in almost half the time, without reducing the generalization performance of the resulting models. We analyze the convergence properties of our multirate scheme and draw a comparison with vanilla SGD. We also discuss splitting choices for the neural network parameters which could enhance generalization performance when neural networks are trained from scratch. A multirate approach can be used to learn different features present in the data and as a form of regularization. Our paper unlocks the potential of using multirate techniques for neural network training and provides several starting points for future work in this area.