[论文笔记]Rethinking the Inception Architecture for Computer Vision
2018-03-15 23:45
645 查看
摘要:
本文探索了一种方法,通过尽可能有效的稳定分解卷积核和正则化来利用增加的计算量来逐级增加网络的规模。这样单个网络可以在ImageNet数据集上达到21.2%的top-1错误率和5.6%的top-5错误率。与此同时,通过级联多个网络,可以达到17.3%的top-1错误率和3.5%的top-5错误率。
本文探索了一种方法,通过尽可能有效的稳定分解卷积核和正则化来利用增加的计算量来逐级增加网络的规模。这样单个网络可以在ImageNet数据集上达到21.2%的top-1错误率和5.6%的top-5错误率。与此同时,通过级联多个网络,可以达到17.3%的top-1错误率和3.5%的top-5错误率。
相关文章推荐
- [深度学习论文笔记][Image Classification] Rethinking the Inception Architecture for Computer Vision
- 论文笔记 | Rethinking the Inception Architecture for Computer Vision
- InceptionV3 Rethinking the Inception Architecture for Computer Vision 论文阅读
- Rethinking the Inception Architecture for Computer Vision 笔记
- Rethinking the Inception Architecture for Computer Vision-Inception-V3
- Rethinking the inception architecture for computer vision的 paper 相关知识
- 论文笔记——Rethinking the Inception Architecture for Computer Vision
- rethinking the inception architecture for computer vision
- Rethinking the Inception Architecture for Computer Vision
- READING NOTE: Rethinking the Inception Architecture for Computer Vision
- 《Rethinking the Inception Architecture for Computer Vision》笔记
- Inception V3(Rethinking the Inception Architecture for Computer Vision)
- Inception-v3:"Rethinking the Inception Architecture for Computer Vision"
- Inception系列2_Rethinking the Inception Architecture for Computer Vision
- 网络模型 Inception V2/V3-Rethinking the Inception Architecture for Computer Vision
- DeepLearning论文笔记(一):What is the Best Muti-Stage Architecture for object Recognition
- 1703.In Defense of the Triplet Loss for Person Re-Identification 论文阅读笔记
- 【论文笔记】In Defense of the Triplet Loss for Person Re-Identification
- 论文阅读笔记:Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey
- 论文阅读笔记:MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications