Reversible Column Networks

TL;DR

Reversible Column Networks (RevCol) introduce a novel neural architecture with reversible connections that maintain total information during forward pass, leading to state-of-the-art results in vision tasks and potential applications in NLP.

Contribution

The paper presents RevCol, a new neural network paradigm with reversible columns that enhance information retention and improve performance across vision and NLP tasks.

Findings

RevCol achieves 88.2% ImageNet-1K accuracy after pre-training.

RevCol outperforms existing CNNs on COCO detection and ADE20k segmentation.

RevCol can be integrated into transformers to boost performance.

Abstract

We propose a new neural network design paradigm Reversible Column Network (RevCol). The main body of RevCol is composed of multiple copies of subnetworks, named columns respectively, between which multi-level reversible connections are employed. Such architectural scheme attributes RevCol very different behavior from conventional networks: during forward propagation, features in RevCol are learned to be gradually disentangled when passing through each column, whose total information is maintained rather than compressed or discarded as other network does. Our experiments suggest that CNN-style RevCol models can achieve very competitive performances on multiple computer vision tasks such as image classification, object detection and semantic segmentation, especially with large parameter budget and large dataset. For example, after ImageNet-22K pre-training, RevCol-XL obtains 88.2% ImageNet-1K accuracy. Given more pre-training data, our largest model RevCol-H reaches 90.0% on ImageNet-1K, 63.8% APbox on COCO detection minival set, 61.0% mIoU on ADE20k segmentation. To our knowledge, it is the best COCO detection and ADE20k segmentation result among pure (static) CNN models. Moreover, as a general macro architecture fashion, RevCol can also be introduced into transformers or other neural networks, which is demonstrated to improve the performances in both computer vision and NLP tasks. We release code and models at https://github.com/megvii-research/RevCol