Resnet in Resnet: Generalizing Residual Architectures
Sasha Targ, Diogo Almeida, Kevin Lyman
TL;DR
This paper introduces Resnet in Resnet (RiR), a dual-stream architecture that generalizes ResNets, improves performance on CIFAR datasets, and sets new state-of-the-art results without additional computational cost.
Contribution
The paper proposes RiR, a novel dual-stream residual architecture that enhances ResNet performance and is easy to implement with no extra computational overhead.
Findings
RiR outperforms ResNet on CIFAR-10 and CIFAR-100.
RiR establishes new state-of-the-art on CIFAR-100.
RiR improves accuracy with no additional computational cost.
Abstract
Residual networks (ResNets) have recently achieved state-of-the-art on challenging computer vision tasks. We introduce Resnet in Resnet (RiR): a deep dual-stream architecture that generalizes ResNets and standard CNNs and is easily implemented with no computational overhead. RiR consistently improves performance over ResNets, outperforms architectures with similar amounts of augmentation on CIFAR-10, and establishes a new state-of-the-art on CIFAR-100.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdvanced Neural Network Applications · Adversarial Robustness in Machine Learning · Domain Adaptation and Few-Shot Learning
MethodsAverage Pooling · *Communicated@Fast*How Do I Communicate to Expedia? · 1x1 Convolution · Batch Normalization · Bottleneck Residual Block · Global Average Pooling · Residual Block · Kaiming Initialization · Max Pooling · Residual Connection