Resnet BN

【深度學習】深刻理解Batch Normalization批標準化

https://www.zhihu.com/topic/20084849/hot

resnet（殘差網絡）的F（x）究竟長什麼樣子？

https://www.zhihu.com/question/53224378

如何理解微軟的深度殘差學習？

https://www.zhihu.com/question/38499534?sort=created

 

SKIP CONNECTIONS ELIMINATE SINGULARITIES

https://arxiv.org/pdf/1701.09175.pdf

 詳解殘差網絡

https://zhuanlan.zhihu.com/p/42706477

 

殘差網絡原理

https://blog.csdn.net/qq_30478885/article/details/78828734

https://www.coursera.org/lecture/convolutional-neural-networks/why-resnets-work-XAKNO

https://arxiv.org/pdf/1512.03385.pdf

https://www.quora.com/How-does-deep-residual-learning-work

https://arxiv.org/pdf/1603.05027.pdf

Resnet中殘差塊的做用是完成恆等變換，那這樣的恆等變換的意義是什麼，在網絡中能起到怎樣的做用呢？

https://www.zhihu.com/question/293243905

https://zhuanlan.zhihu.com/p/28124810 

https://arxiv.org/pdf/1502.03167v3.pdf

https://zhuanlan.zhihu.com/p/31645196

https://www.coursera.org/lecture/convolutional-neural-networks/why-resnets-work-XAKNO

https://arxiv.org/pdf/1506.01497v3.pdf

https://arxiv.org/pdf/1504.08083.pdf

https://arxiv.org/pdf/1311.2524v5.pdf

https://arxiv.org/pdf/1702.08591.pdf

https://arxiv.org/pdf/1611.05431.pdf

https://arxiv.org/pdf/1607.07032.pdf

 

https://arxiv.org/abs/1605.06431

Residual Networks 理解

協方差

https://www.zhihu.com/question/20852004

ResNet架構可逆！多大等提出性能優越的可逆殘差網絡

一文簡述ResNet及其多種變體

 

TensorFlow 實現 Resnet V2 代碼解讀

Identity Mapping in ResNet

1. 學習吳恩達在coursera的「深度學習課程」中關於殘差網絡的內容
2. 讀該模型的原版論文：Deep Residual Learning for Image Recognition，若是閱讀有難度，能夠參考網絡上的翻譯稿，這裏有一篇筆者的翻譯稿供參考。
3. 註冊github，用於查看和下載殘差網絡的開源源碼。註冊地址。
4. 複製源代碼到本地。源碼地址在此。

 

【1】He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 770-778.

【2】Srivastava R K, Greff K, Schmidhuber J. Highway networks[J]. arXiv preprint arXiv:1505.00387, 2015.

【3】Orhan A E, Pitkow X. Skip connections eliminate singularities[J]. arXiv preprint arXiv:1701.09175, 2017.

【4】Shang W, Sohn K, Almeida D, et al. Understanding and Improving Convolutional Neural Networks via Concatenated Rectified Linear Units[J]. 2016:2217-2225.

【5】Greff K, Srivastava R K, Schmidhuber J. Highway and Residual Networks learn Unrolled Iterative Estimation[J]. 2017.

【6】Jastrzebski S, Arpit D, Ballas N, et al. Residual connections encourage iterative inference[J]. arXiv preprint arXiv:1710.04773, 2017.