Scribd Logo
Upload

Welcome to Scribd!

  • ANN - Session 2 CO 1

    Uploaded by

    joxekoj
    7 views9 pages
    The document summarizes three different neuron models: 1) McCulloch Pitts Model - A basic model where activation is a weighted sum of inputs and the output is a nonlinear function. Weights are fixed so it cannot learn. 2) Rosenblatt Perceptron Model - An improvement over McCulloch Pitts that can process non-boolean inputs, assign different weights, and automatically compute thresholds. It also incorporates learning by adjusting weights based on error. 3) Adaptive Linear Element (ADALINE) - A model that sets initial random weights and bias, then calculates net input and updates weights and bias iteratively to minimize mean squared error between target and actual values.

    Original Description:

    Original Title

    ANN_Session-2-CO-1

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    The document summarizes three different neuron models: 1) McCulloch Pitts Model - A basic model where activation is a weighted sum of inputs and the output is a nonlinear function. Weights are fixed so it cannot learn. 2) Rosenblatt Perceptron Model - An improvement over McCulloch Pitts that can process non-boolean inputs, assign different weights, and automatically compute thresholds. It also incorporates learning by adjusting weights based on error. 3) Adaptive Linear Element (ADALINE) - A model that sets initial random weights and bias, then calculates net input and updates weights and bias iteratively to minimize mean squared error between target and actual values.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Flag for inappropriate content
    7 views9 pages

    ANN - Session 2 CO 1

    Uploaded by

    joxekoj
    The document summarizes three different neuron models: 1) McCulloch Pitts Model - A basic model where activation is a weighted sum of inputs and the output is a nonlinear function. Weights are fixed so it cannot learn. 2) Rosenblatt Perceptron Model - An improvement over McCulloch Pitts that can process non-boolean inputs, assign different weights, and automatically compute thresholds. It also incorporates learning by adjusting weights based on error. 3) Adaptive Linear Element (ADALINE) - A model that sets initial random weights and bias, then calculates net input and updates weights and bias iteratively to minimize mean squared error between target and actual values.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Flag for inappropriate content
    of 9

    Neuron Model

    ●
    McCulloch Pitts Model Rosenblatt
    ●
    perceptron model
    ●

    Adaptive Linear element (ADALINE)


    1.5.1 McCulloch Pitts Model
    ● In McCulloch-Pitts (MP) model the activation (x) is given by
    a weighted sum of its M input values (ai) and a bias term.
    ● The output signal (s) is typically a nonlinear function f(x) of the
    activation value x.
    ●
    Networks consisting of MP neurons with binary (on-off) output
    signals can be configured to perform several logical functions.
    ●
    In the MP model the weights are fixed. Hence a network using
    this model does not have the capability of learning.
    McCulloch Pitts Model

    The following equations describe the operation of an MP model:


    McCulloch Pitts Model

    The following equations describe the operation of an MP model:


    1.5.2 Rosenblatt Perceptron Model
    ●
    It was designed by Rosenblatt in 1958 to overcome most of the issues of the
    McCulloch Pitts model.
    1. It can process non boolean inputs
    2. It can assign different weights to each input.
    ●
    3. The threshold is computed automatically.
    The Rosenblatt's perceptron model for an artificial neuron consists of outputs
    ● from sensory units to a fixed set of association units the outputs of which are
    fed to an MP neuron.
    The main deviation from the M P model is that learning(i.e., adjustment of
    weights) is incorporated in the operation of the unit. The desired or target
    output (b) is compared with the actual binary output (s), and the error (e) is
    used to adjust the weights.
    Rosenblatt Perceptron Model
    Rosenblatt Perceptron Model
    1.5.3 Adaptive Linear element
    Algorithm: (ADALINE)
    ●
    Weight and bias are set to some random values
    ● but not zero.
    Calculate the net input to output unit untill the
    ●
    least mean sqared error is obtained.
    Error= target value – activated value update the
    weight and bias for i=1 to n.
    Adaptive Linear element
    (ADALINE)

    You might also like