Cyan's Blog

Search

Search IconIcon to open search

D2L-43-GoogLeNet(Inception)

Last updated Mar 5, 2022 Edit Source

# GoogLeNet

2022-03-05

Tags: #DeepLearning #CNN #GoogLeNet-Inception

# Inception Block

# 并行连结: 不同卷积核的组合

# Add Sparsity using Dense Building Blocks

# 1x1卷积: 减少通道的数量

# Computational Efficiency

3

# GoogLeNet

200

# Auxiliary Networks: Not Important

# Future Improvements

Inception变种


  1. S. Arora, A. Bhaskara, R. Ge, and T. Ma, “Provable bounds for learning some deep representations,” in International conference on machine learning, 2014, pp. 584–592. PDF(zotero://select/items/@arora2014provable) ↩︎

  2. knowyourmeme.com/memes/we-need-to-go-deeper 电影中有这么一句话: “We need to go deeper”。 ↩︎

  3. 27 含并行连结的网络 GoogLeNet / Inception V3【动手学深度学习v2】哔哩哔哩_bilibili ↩︎

  4. GoogLeNet系列解读_shuzfan的专栏-CSDN博客_googlenet ↩︎