Learning with Rethinking: Recurrently Improving Convolutional Neural Networks through Feedback

TL;DR

This paper introduces a feedback mechanism in CNNs that allows models to iteratively refine their predictions, enhancing performance on various image classification benchmarks.

Contribution

It proposes a novel feedback layer and emphasis vector for recurrently improving CNNs, applicable to pre-trained models, which is a new approach in CNN training.

Findings

Improved accuracy on CIFAR-100, CIFAR-10, MNIST-background-image, and ILSVRC-2012 datasets.

Feedback mechanism enhances CNN performance through iterative refinement.

Applicable to existing pre-trained models for performance boost.

Abstract

Recent years have witnessed the great success of convolutional neural network (CNN) based models in the field of computer vision. CNN is able to learn hierarchically abstracted features from images in an end-to-end training manner. However, most of the existing CNN models only learn features through a feedforward structure and no feedback information from top to bottom layers is exploited to enable the networks to refine themselves. In this paper, we propose a "Learning with Rethinking" algorithm. By adding a feedback layer and producing the emphasis vector, the model is able to recurrently boost the performance based on previous prediction. Particularly, it can be employed to boost any pre-trained models. This algorithm is tested on four object classification benchmark datasets: CIFAR-100, CIFAR-10, MNIST-background-image and ILSVRC-2012 dataset. These results have demonstrated the advantage of training CNN models with the proposed feedback mechanism.