GuideBP: Guiding Backpropagation Through Weaker Pathways of Parallel Logits

TL;DR

GuideBP introduces a novel method for guiding backpropagation along weaker pathways in parallel CNN logits, improving training efficiency and model performance in multi-branch architectures.

Contribution

It proposes a class-specific weakness scoring mechanism to direct gradients along weaker pathways, enhancing multi-branch CNN training and performance.

Findings

Outperforms traditional logit merging techniques

Improves training efficiency for multiple model instances

Enhances CNN performance with multiple output branches

Abstract

Convolutional neural networks often generate multiple logits and use simple techniques like addition or averaging for loss computation. But this allows gradients to be distributed equally among all paths. The proposed approach guides the gradients of backpropagation along weakest concept representations. A weakness scores defines the class specific performance of individual pathways which is then used to create a logit that would guide gradients along the weakest pathways. The proposed approach has been shown to perform better than traditional column merging techniques and can be used in several application scenarios. Not only can the proposed model be used as an efficient technique for training multiple instances of a model parallely, but also CNNs with multiple output branches have been shown to perform better with the proposed upgrade. Various experiments establish the flexibility of the learning technique which is simple yet effective in various multi-objective scenarios both empirically and statistically.