Neural Networks Multilayer Perceptron MLP Oscar Herrera Alcntara
- Slides: 14
Neural Networks Multilayer Perceptron (MLP) Oscar Herrera Alcántara heoscar@yahoo. com
Outline n Neuron n n Artificial neural networks Activation functions n Perceptrons n Multilayer perceptrons n Backpropagation Generalization n Introduction to Artificial Intelligence APSU
Neuron n A neuron is a cell in the brain q q q collection, processing, and dissemination of electrical signals neurons of > 20 types, synapses, 1 ms-10 ms cycle time brain’s information processing relies on networks of such neurons Introduction to Artificial Intelligence APSU
Biological Motivation n dendrites: nerve fibres carrying electrical signals to the cell n cell body: computes a non-linear function of its inputs n axon: single long fiber that carries the electrical signal from the cell body to other neurons n synapse: the point of contact between the axon of one cell and the dendrite of another, regulating a chemical connection whose strength affects the input to the cell. Introduction to Artificial Intelligence APSU
Artificial neural networks n n A mathematical model of the neuron is Mc. Culloch-Pitts unit Neural networks consists of nodes (units) connected by directed links 1 x 1 b : Bias wi 1 x 2 Neuron i S v j y x 3 xm n Wim Synaptic Induced local field Activation Inputs Weights Activation potential function Output A bias weight Wi, 0 connected to a fixed input xi, 0 = +1 Introduction to Artificial Intelligence APSU
Activation functions a) Step function or Threshold function b) Sigmoid function c) Hyperbolic tangent function Introduction to Artificial Intelligence APSU
Perceptron learning n Learn by adjusting weights to reduce error on training set q Error correction learning rule q Perform optimization search by gradient descent Introduction to Artificial Intelligence APSU
Implementing logic functions n Mc. Culloch-Pitts unit can implement any Boolean function Introduction to Artificial Intelligence APSU
Expressiveness of perceptrons n A perceptron q q can represent AND, OR, NOT can represent a linear separator (function) in input space: Introduction to Artificial Intelligence APSU
Multilayer Perceptron (MLP): Architecture Bias Input Hidden Layers Layer j j x 1 Inputs x 2 x 3 1 j Output Layer j j y 1 j y 2 1 j 1 wij j j wjk j Introduction to Artificial Intelligence APSU wkl Outputs
Solve XOR problem using MLPs n n n A two-layer network with two nodes in the hidden layer The hidden layer maps the points from non linear separable space to linear separable space. The output layer finds a decision line j (v) Introduction to Artificial Intelligence APSU
Back-propagation Algorithm 1. Initialization. Weights are initialized with random values whose mean is zero 2. Presentations of training examples 3. Forward computation 4. -Backward computation for the neuron j of the hidden layer l for the neuron j of the output layer L 5. - Iteration. Repeat step 2 to 4 until E< desired error a the momentum parameter is ajusted h the learning-rate parameter is ajusted Introduction to Artificial Intelligence APSU
MLP Training i Left j k Forward Pass • Fix wji(n) • Compute yj(n) x Right y Backward Pass • Calculate dj(n) • Update weights wji(n+1) Left i j Introduction to Artificial Intelligence APSU k Right
Generalization n Total Data are divided in two parts: q q n n Data Training (80%) MLP is trained with Data Training Data Test (20%) MLP is tested with Data Test Generalization MLP is used with inputs which have never been presented in order to predict the outputs Introduction to Artificial Intelligence APSU
- Multilayer perceptron calculation example
- Multilayer perceptron
- Bagdt
- Miguel herrera maritza herrera
- Multilayer neural network
- Building cisco multilayer switched networks
- 11-747 neural networks for nlp
- Style transfer
- Neural networks and learning machines
- Threshold logic unit
- Lstm components
- Few shot learning with graph neural networks
- Freed et al 2001 ib psychology
- Neuraltools neural networks
- Perceptron xor