1 / 16

Neural Networks

Neural Networks. Dr. Thompson March 19, 2013. Artificial Intelligence. Robotics Computer Vision & Speech Recognition Expert Systems Pattern Recognition Machine Learning Natural Language Processing Prognostics & Diagnostics. Neural Network Applications. Character Recognition

clea
Download Presentation

Neural Networks

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Neural Networks Dr. Thompson March 19, 2013

  2. Artificial Intelligence • Robotics • Computer Vision & Speech Recognition • Expert Systems • Pattern Recognition • Machine Learning • Natural Language Processing • Prognostics & Diagnostics

  3. Neural Network Applications • Character Recognition • Loan Officer • Cancer Diagnosis • Wine Classifier • Stock Market Prediction • Network Security

  4. HRL • Artificial Neural Networks - Pattern Recognition • Airbag Problem – Accelerometer False Positive • OCR Check Character Recognition • Bayesian Networks – Expert System • GM Electromotive Division • Amazon.com Buyer Preferences

  5. Biological Neuron

  6. Artificial Neural Network Topology

  7. Artificial Neuron Activation

  8. Threshhold Functions(include graphs) • Linear • Logistic • Hyperbolic Tangent – Sigmoid (*) • Step Logistic Curve

  9. Network Output • Y = f(WX) • Z = f(W’Y) = f(W’f(WX))

  10. Error Correction • (Method of Least Squares) • Minimize Total Error = E = Σ(Z-O)2

  11. Solution Space

  12. Error Function:Local & Global Minima

  13. Back PropagationDelta Rule – Gradient Descent • http://en.wikipedia.org/wiki/Delta_Rule

  14. Learning & Testing • Heuristics • 10% • 90% • Overtraining/Overfitting • Polynomial Curve Fitting Analogy

  15. Next Week Matlab Neural Network Toolbox Tutorial

  16. Assignment • Read the Wikipedia Artificial Neural Network & Backpropagation Chapters • Devise a Neural Network Characterization of • 4x4 Scoreboard Digit Problem • Input Layer • Hidden Layer • Output Layer • Training Set Examples

More Related