Download presentation
Presentation is loading. Please wait.
Published byJohan Tedjo Modified over 6 years ago
1
NEURAL NETWORKS Lab 1 dr Zoran Ševarac FON, 2013.
2
WHAT IS NEURAL NETWORK Brain inspired mathematical models
Biological and artificial NN
3
Biological and artificial neuron
Basic parts: body(soma), dendrites(inputs), axons (outputs), synapses (connections)
4
SUPERVISED LEARING Neural network
Basic principle: iterative error minimization Desired output d(x) Error Neural network Input X Actual output y(x) e(x) = d(x) – y(x)
5
ADALINE Linearna transfer function Linear combination of inputs
w1 w2 w3 y Linearna transfer function Linear combination of inputs y = w1u1+w2u2+…+wnun, Learning algorithm: Least Mean Squares
6
PERCEPTRON Step transfer function
y1 y2 y3 Step transfer function Perceptron learning - first algorithm for learning nonlinear systems Only for linear separable problems
7
Multi Layer Perceptron
Basic perceptrn extended with one or more layers of hidden neurons between input and output layer Differentiable neuron transfer functions (tanh, sigmoid) Using Backpropagation algorithm for learning, which is bassed on LMS algorithm Able to solve complex problems
8
NEUROPH FRAMEWORK
9
Osnovni koncepti
11
Neuroph – learning rules class diagram
12
Creating and training neural networks in Java
// create new perceptron network NeuralNetwork neuralNetwork = new Perceptron(2, 1, TransferFunction.STEP); // create training set DataSet trainingSet=new DataSet(2, 1); // add training data to training set trainingSet.addRow(new DataSetRow(“0 0”, “0”)); trainingSet. addRow(new DataSetRow(“0 1”, “1”)); trainingSet. addRow(new DataSetRow(“1 0”, “1”)); trainingSet. addRow(new DataSetRow(“1 1”, “1”)); // learn the training set neuralNetwork.learn(trainingSet); // save the trained network into file neuralNetwork.save(“or_perceptron.nnet”);
13
Demo Neuroph Studio: learning basic logic functions
14
LINKS AND BOOKS http://is.fon.rs/neuronske_mreze
Neural Networks - A Systematic Introduction , besplatna online knjiga Introduction to Neural Computation
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.