Professional Documents
Culture Documents
Lecture03 VDL
Lecture03 VDL
Lecture 04
◼ AlexNet
input image
weights
loss
input image
loss
◼ A simple example
◼ Chain Rule
◼ A simple example
Forward pass
Backward pass
◼ A simple example
Forward pass
Backward pass
2 3𝑥 2 1 1*2
2 2
6𝑥 2 2 1
◼ As a neural network
◼ Gradient Descent
^ Updates after looking at complete dataset
◼ Minibatch Gradient Descent
^ Updates after looking at N samples (batch size)
◼ Stochastic Gradient Descent
^ Updates after looking at every samples
◼ Related Concept
^ Epoch
• one cycle through the full training dataset
◼ Linear Classifier
𝑥1
ሺ 1 1ሻ > 0.5
ด
𝑇
𝑥2 −𝑤
𝒙𝟏 𝒙𝟐 OR (𝒙𝟏 , 𝒙𝟐 ) 𝒘 0
𝒙
0 0 0
0 1 1
1 0 1
1 1 1
◼ Linear Classifier 𝑥1
ሺ 1 1ሻ > 1.5
ด
𝑇
𝑥2 −𝑤
𝒘 0
𝒙
𝒙𝟏 𝒙𝟐 AND (𝒙𝟏 , 𝒙𝟐 )
0 0 0
0 1 0
1 0 0
1 1 1
◼ Linear Classifier 𝑥1
ሺ−1 −1ሻ > − 1.5
ด
𝑇
𝑥2 −𝑤
𝒘 0
𝒙
𝒙𝟏 𝒙𝟐 NAND (𝒙𝟏 , 𝒙𝟐 )
0 0 1
0 1 1
1 0 1
1 1 0
◼ Linear Classifier
𝒙𝟏 𝒙𝟐 XOR (𝒙𝟏 , 𝒙𝟐 )
0 0 0
0 1 1
1 0 1
1 1 0
Rosenblatt: The perceptron -a probabilistic model for information storage and organization in the brain. Psychological Review, 1958.
Minsky and Papert: Perceptrons: An introduction to computational geometry. MIT Press, 1969.
◼ Linear Classifier
𝒙𝟏 𝒙𝟐 XOR (𝒙𝟏 , 𝒙𝟐 )
0 0 0
0 1 1
1 0 1
1 1 0
◼ https://playground.tensorflow.org/
◼ Neuron
^ Electrically excitable cell that communicates
with other cells via specialized connections
called synapses (100 billion)
◼ Sensory neurons
^ 5 senses
◼ Motor neurons
^ Allow brain to communicate with other parts
of the body
◼ Interneurons
^ connect neurons to other neurons within the
same region of the brain
◼ Problems
^ Saturates: The gradients are
killed
^ Outputs are not zero centred
◼ Problems
Restricts gradient updates and is the reason for inefficient optimisation(minibatch helps)
◼ Zero centred
◼ Antisymmetric
◼ Problem
^ Saturation Kills the gradient
Rumelhart, Hinton and Williams: Learning representations by back-propagating errors. Nature, 1986.
◼ Concepts
^ Gradient Descent
• Vanilla, Minibatch, Stochastic
◼ Simple Functions
^ OR, AND, NAND, XOR
◼ Representation Matters
◼ Neural Network Playground
◼ MLP
◼ Activation Functions