W3.Neural Network Representation

You might also like

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 7

One hidden layer

Neural Network

Neural Network
deeplearning.ai
Representation
Neural Network Representation

 
𝑥1
 
𝑥2  
^𝑦
 
𝑥3

Andrew Ng
One hidden layer
Neural Network

Computing a
deeplearning.ai Neural Network’s
Output
Neural Network Representation

 𝑥
1   𝑥1
 𝑥
2   𝑤 𝑇 𝜎(𝑧)
𝑥 +𝑏  𝑎=^𝑦
    𝑥2 𝑦
 ^

𝑧
  𝑎
 
  𝑥3
 𝑥
3

𝑧=𝑤𝑇 𝑥 +𝑏
 

𝑎=𝜎 (𝑧)
 

Andrew Ng
Neural Network Representation

 𝑥
1   𝑥1
 𝑥
2   𝑤 𝑇 𝜎(𝑧)
𝑥 +𝑏  𝑎=^𝑦
    𝑥2 ^ 𝑦
𝑧
  𝑎
 
  𝑥3
 𝑥
3

𝑧=𝑤𝑇 𝑥 +𝑏
    𝑥1
𝑎=𝜎 (𝑧)
    𝑥2 ^ 𝑦
  𝑥3
Andrew Ng
Neural Network Representation
 𝑎 [1 ]
1
  )
 
𝑥1  𝑎 [1 ]   )
2
  𝑥2  𝑎 [1 ]
𝑦
 ^
  )
3
 
𝑥3
 𝑎 [1 ]   )
4

Andrew Ng
Neural Network Representation learning
 𝑎 [1 ]
1
Given input x:
 
𝑥1  𝑎 [1 ]
2   [ 1]
𝑧 =𝑊 [ 1 ] 𝑥+𝑏 [ 1]
  𝑥2  𝑎 [1 ]
𝑦
 ^
3
 
𝑥3  𝑎 [1 ] =𝜎 (𝑧 [1 ] )
 𝑎 [1 ]
4

  [ 2]
𝑧 =𝑊 [ 2 ] 𝑎 [1 ] +𝑏 [ 2]

 𝑎 [2 ] =𝜎 (𝑧 [2 ] )

Andrew Ng

You might also like