ResNet: Enabling Deep Convolutional Neural Networks through Residual Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908617015296000 |
|---|---|
| author | Liu, Xingyu Goh, Kun Ming |
| author_facet | Liu, Xingyu Goh, Kun Ming |
| contents | Convolutional Neural Networks (CNNs) has revolutionized computer vision, but training very deep networks has been challenging due to the vanishing gradient problem. This paper explores Residual Networks (ResNet), introduced by He et al. (2015), which overcomes this limitation by using skip connections. ResNet enables the training of networks with hundreds of layers by allowing gradients to flow directly through shortcut connections that bypass intermediate layers. In our implementation on the CIFAR-10 dataset, ResNet-18 achieves 89.9% accuracy compared to 84.1% for a traditional deep CNN of similar depth, while also converging faster and training more stably. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_24036 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ResNet: Enabling Deep Convolutional Neural Networks through Residual Learning Liu, Xingyu Goh, Kun Ming Computer Vision and Pattern Recognition Artificial Intelligence Convolutional Neural Networks (CNNs) has revolutionized computer vision, but training very deep networks has been challenging due to the vanishing gradient problem. This paper explores Residual Networks (ResNet), introduced by He et al. (2015), which overcomes this limitation by using skip connections. ResNet enables the training of networks with hundreds of layers by allowing gradients to flow directly through shortcut connections that bypass intermediate layers. In our implementation on the CIFAR-10 dataset, ResNet-18 achieves 89.9% accuracy compared to 84.1% for a traditional deep CNN of similar depth, while also converging faster and training more stably. |
| title | ResNet: Enabling Deep Convolutional Neural Networks through Residual Learning |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2510.24036 |