Impact of Aliasing on Generalization in Deep Convolutional Networks
원문 초록
We investigate the impact of aliasing on generalization in Deep Convolutional Networks and show that data augmentation schemes alone are unable to prevent it due to structural limitations in widely used architectures. Drawing insights from frequency analysis theory, we take a closer look at ResNet and EfficientNet architectures and review the trade-off between aliasing and information loss in each of their major components. We show how to mitigate aliasing by inserting non-trainable low-pass filters at key locations, particularly where networks lack the capacity to learn them. These simple architectural changes lead to substantial improvements in generalization on i.i.d. and even more on out-of-distribution conditions, such as image classification under natural corruptions on ImageNet-C [11] and few-shot learning on Meta-Dataset [26]. State-of-the art results are achieved on both datasets without introducing additional trainable parameters and using the default hyper-parameters of open source codebases.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
FewVS: A Vision-Semantics Integration Framework for Few-Shot Image Classification
IPCL: Iterative Pseudo-Supervised Contrastive Learning to Improve Self-Supervised Feature Representation
Multistage Relation Network With Dual-Metric for Few-Shot Hyperspectral Image Classification
DGSSC: A Deep Generative Spectral-Spatial Classifier for Imbalanced Hyperspectral Imagery
Few-shot Learning with Class-Covariance Metric for Hyperspectral Image Classification
DocXClassifier: High Performance Explainable Deep Network for Document Image Classification
토론
이 논문과 연결된 포럼 글이 아직 없습니다.