Neural Rejuvenation: Improving Deep Network Training by Enhancing Computational Resource Utilization

TL;DR

This paper introduces Neural Rejuvenation, a novel optimization technique that enhances neural network training by reallocating resources to dead neurons, significantly improving performance without increasing resource usage.

Contribution

The paper proposes Neural Rejuvenation, a new method for detecting and revitalizing dead neurons to better utilize computational resources during training.

Findings

Significant performance improvements across various neural networks.

Effective detection and rejuvenation of dead neurons in real time.

Maintains similar resource usage while boosting accuracy.

Abstract

In this paper, we study the problem of improving computational resource utilization of neural networks. Deep neural networks are usually over-parameterized for their tasks in order to achieve good performances, thus are likely to have underutilized computational resources. This observation motivates a lot of research topics, e.g. network pruning, architecture search, etc. As models with higher computational costs (e.g. more parameters or more computations) usually have better performances, we study the problem of improving the resource utilization of neural networks so that their potentials can be further realized. To this end, we propose a novel optimization method named Neural Rejuvenation. As its name suggests, our method detects dead neurons and computes resource utilization in real time, rejuvenates dead neurons by resource reallocation and reinitialization, and trains them with new training schemes. By simply replacing standard optimizers with Neural Rejuvenation, we are able to improve the performances of neural networks by a very large margin while using similar training efforts and maintaining their original resource usages.