mlatoz
CNN Intuition
Plan Of Attack
What we will learn in this section:
- What are Convolutional Neural Networks?
- Step 1 - Convolution Operation
- Step 1 (b) - ReLU (Rectified Linear Unit) Layer
- Step 2 - Pooling
- Step 3 - Flattening
- Step 4 - Full Connection
- Summary
What are Convolutional Neural Networks?
How does they work?
Example
- STEP 1: Convolution
- STEP 2: Max Pooling
- STEP 3: Flattening
- STEP 4: Full Connection
Additional Reading
Gradient-Based Learning Applied to Document Recognition
By Yann LeCun et al. (1998)
Gradient-Based Learning Applied to Document Recognition - Yann LeCun
Step 1 - Convolution Operation
(f * g)(t) =def -∞∫∞f(Ƭ) g(t - Ƭ) dƬ
- A Convolution is basically a combined integration of two functions, and it shows you how one function modifies the other.
Additional Reading
Introduction to Convolutional Neural Networks
By Jianxin Wu (2017)
Introduction to Convolutional Neural Networks - Jianxin Wu
Step 1(B) - ReLU (Rectified Linear Unit) Layer
Additional Reading
Understanding Convolutional Neural Networks with A Mathematical Model
By C. C. Jay Kuo (2016)
Understanding Convolutional Neural Networks with A Mathematical Model - C. C. Jay Kuo
Additional Reading
Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
By Kaiming He et al. (2015)
Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification - Kaiming He
Step 2 - Max Pooling
Additional Reading
Evaluation of Pooling Operations in Convolutional Architectures for Object Recognition
By Dominik Scherer et al. (2010)
Evaluation of Pooling Operations in Convolutional Architectures for Object Recognition - Dominik Scherer
Step 3 - Flattening
Step 4 - Full Connection
Summary
Additional Reading
The 9 Deep Learning Papers You Need To Know About (Understanding CNNs Part 3)
By Adit Deshpande (2016)
The 9 Deep Learning Papers You Need To Know About (Understanding CNNs Part 3) - Adit Deshpande
Soft-Max & Cross-Entropy
- Soft-Max Function
fj(z) = ezj / ∑k ezk
- Cross-Entropy Function
Li = -log(efyi / ∑j efj)
H(p, q) = -∑x p(x) log q(x)
Additional Reading
A Friendly Introduction to Cross-Entropy Loss
By Rob DiPietro (2016)
A Friendly Introduction to Cross-Entropy Loss - Rob DiPietro
Additional Reading
How to Implement a Neural Network Intermezzo 2
By Peter Roelants (2016)
How to Implement a Neural Network Intermezzo 2 - Peter Roelants
| «Previous |