1 / 7

Explorations in Neural Networks

Explorations in Neural Networks. Tianhui Cai Period 3. Definition. Mathematical model based on biological neural networks Interconnected group of artificial neurons Highly connected Each unit is simple, but the system is complex

randyevans
Télécharger la présentation

Explorations in 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. Explorations in Neural Networks • Tianhui Cai • Period 3

  2. Definition • Mathematical model based on biological neural networks • Interconnected group of artificial neurons • Highly connected • Each unit is simple, but the system is complex • Neurons have output value, which is determined by the outputs of other nodes that feed into it • Nodes are connected by directed links with weights • Adaptive system – changes structure • Can model complicated functions

  3. Neural Networks

  4. Feedforward networks • Information moves in one direction – forward • Input -> hidden -> output • Single layer perceptron • Consists of a single layer of output nodes • Inputs are fed directly to output with weights • Sum of weights * inputs is calculated for a node • Neuron 'fires' based on activation function • Limited functionality. • Multi-layer perceptron • Multiple layers • Learns through back-propagation

  5. Applications • Pattern recognition • Classification • Handwritten digit classification • Useful for reading zip codes • Can deal with noisy samples

  6. Current status • Can make a neural network • Feedforward • Multiple layers • Neural networks can take input and spit out output correctly • Backpropagation • Binary data – AND, OR, XOR

  7. Future • Gather input data for handwriting samples • Alter program to process handwriting samples / images • more inputs • more outputs • Test variations • Number of hidden layers • Number of hidden neurons • Specific connections

More Related