Jonathan Reagan Umass Dartmouth CSUMS Summer 11 August 3 rd 2011
What is a Neural Network? How does it work? Why do we care? Results Issues encountered Future work
Input Layers Hidden Layers Output Layers
Not realistic to study every possible case Smaller sample can be used to model the entire case Assume connections hold
(input)=[age, income, credit score, etc] (output)=[dependability] We want weights of α’s X* α (Hidden)=Y
Use the learning method to find α I Y-Xα I=0
PerceptronLeast Square NAccuracys Failed TrialsNAccuracys Failed Trials MinAVGMaxN/Total MinAVGMaxN/Total / / / / / / / / / / / / / / / / / /100
2 million Convergence Failed Trials 4 million convergence Failed Trials NT(Time)SecondsN/TotalNT(Time)SecondsN/Total / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /50
Random Data can’t be learned Deterministic Data can be learned Adding Random variance decreases Accuracy More values of N the Better But more values of N take Longer
Increase the speed of the Neural Network Find more applicable data for testing of the Neural Network Try multiple layer Neural Networks and Compare