Concept & Implementation of Perceptron

before we go …

Reference link

Perceptron: https://en.wikipedia.org/wiki/Perceptron
Single Layer Perceptron: http://untitledtblog.tistory.com/27
Multilayer Perceptron: http://untitledtblog.tistory.com/35

Let’s start by understanding these contents.

In Professor Ian’s Data Mining textbook, perceptron is defined as follows.

The resulting polynomial ax ^ 2 + bx ^ 2 + cx ^ 2 + dx ^ 2 + ex ^ 2 + fx ^ 2 …

Is defined as a set of hyperplanes.

In the neural network, we can express the weight for each cell and its connection.

Since neural networks have begun with human neuron simulations, single artificial neurons

  • Threshold – threshhold – Minimum input value that causes nerve reaction
  • Weight – weight – weight for neuron response
  • Neuron – neuron itself
    (Such as nerves, specific input – tactile, nociceptive, etc.)
  • Interception of bias-linear boundary
  • Net value – sum of input values ​​and weights, which is the sum of the input values ​​of a single cell with multiple synapses.
  • Active function – a function that simulates the behavior of a neuron
    After receiving the input for each set of neurons, multiplying the weights for these inputs to the neuron, processing the values ​​in the active function,

When children learn the language, if they say the right words, the parents respond positively, and if they say the wrong words, they go through the process of correcting them with the negative words – language learning.

657 total views, 5 views today

Leave a Reply