Professional Documents
Culture Documents
Andrew French
This week
1
2
-1
-4
0
5 Neural
-6
Network “Not a root tip”
-4
2
1
-3
…
Ideally we want to learn Learning algorithm/
4
which features to use to
→Deep learning classifier goes here
What is deep learning?
• Deep learning is a popular AI technique
• Essentially it is a kind of neural network (as we will see)
1
1
• Waste of resources
• Spatial correlation is local
• most of the weights would be 0
1
5
Does this look familiar?
Convolution Layers
• We can afford to learn multiple filters
• e.g. 100 10x10 masks is only 10K
parameters
2
0
Activation functions
Convolutional Neural Network- Activation
✓ Transfer the range to ✓ Zero centred range: [-1, 1] ✓ Solves ✓ Overcome the ‘dying
[0, 1] x Saturate and kill gradients: vanishing/exploding neuron’ problem
x Saturate and kill small gradient at region of gradients x Performance not
gradients: small 0 and 1 ✓ Simple to calculate consistent
gradient at region of 0 x Some neurons can be
and 1 ‘dead’ with negative input
2
1
Fully Convolutional
connected Neural Network – Fully Connected Layer & Softmax
layer and softmax
• Global feature learning: fully connected to all activations in the previous layer, as the same in MLP.
• Softmax - converts the prediction to the range of [0..1] for each class
2
2
Softmax?
• With Linear regression, we try to predict a real value y, for an input value x
• Perhaps we are predicting someone’s age from a picture
• Linear regression is not good if we are predicting if input data should be
assigned class A or class B
• Perhaps we are predicting Happy or not-Happy from a picture
• To do this there are more suitable
kinds of regression
• E.g. We can use Logistic regression
• This can be used to “squash” a
number into one of two classes
• Example: sigmoid function →
Softmax?
• So why do we need a softmax function?
Class Probability
Happy 0.05
Sad 0.32
Grumpy 0.44
Sleepy 0.19
CNNS
What are they doing, and what can they do?
CNN Architectures (Classification)
Biologically inspired
Beyond Classification: Segmentation
Beyond Classification: Stereo
1
2
-1 • Custom features let
Input Image -4 us:
0
5 • Choose useful
-6 features like
-4
2
orientation
1 • Reduce the number
Orientation -3 of variables
…
Features 4
Feature Vector
Machine Learning
1
2
-1
-4
0
5 Neural
-6
Network “Not a root tip”
-4
2
1
-3
…
4 Learning algorithm/
classifier goes here
Convolutional Neural Networks
Our next network was based on the hourglass architecture [1], which we
have adapted to perform simultaneous classification of wheat awns
[1] Newell, Alejandro, Kaiyu Yang, and Jia Deng. "Stacked Hourglass Networks for Human Pose Estimation." European
Conference on Computer Vision. Springer International Publishing, 2016.
Training
!
Conclusion
• CNNs are dominating current vision research
• end-to-end learned solutions
• system components (e.g. learned feature detection in tracking)