Self-Reorganizing and Rejuvenating CNNs for Increasing Model Capacity Utilization
This work addresses the issue of inefficient network capacity utilization for deep learning practitioners, offering an incremental improvement by enhancing existing architectures without modifications.
The paper tackles the problem of low computational resource utilization in convolutional neural networks by introducing a biologically inspired method that reorganizes and rejuvenates parameters to merge redundant neurons and increase capacity utilization, resulting in improved baseline network performance without structural changes, as demonstrated through experimental results showing model-agnostic applicability and performance gains.
In this paper, we propose self-reorganizing and rejuvenating convolutional neural networks; a biologically inspired method for improving the computational resource utilization of neural networks. The proposed method utilizes the channel activations of a convolution layer in order to reorganize that layers parameters. The reorganized parameters are clustered to avoid parameter redundancies. As such, redundant neurons with similar activations are merged leaving room for the remaining parameters to rejuvenate. The rejuvenated parameters learn different features to supplement those learned by the reorganized surviving parameters. As a result, the network capacity utilization increases improving the baseline network performance without any changes to the network structure. The proposed method can be applied to various network architectures during the training stage, or applied to a pre-trained model improving its performance. Experimental results showed that the proposed method is model-agnostic and can be applied to any backbone architecture increasing its performance due to the elevated utilization of the network capacity.