Scribd Logo
Upload

Welcome to Scribd!

  • Back Propagation

    Uploaded by

    Jafer
    8 views19 pages
    Backpropagation is a fundamental concept in training neural networks and deep learning models. It is a supervised learning algorithm that updates the weights and biases of a neural network to minimize the error between the predicted and actual outputs. The backpropagation algorithm works by computing the gradient of the loss function with respect to each weight via the chain rule, computing the gradient layer by layer from the last to the first to avoid redundant computation. It involves feeding inputs forward, calculating errors backward, and updating weights. An example shows initializing weights randomly, doing a forward pass, calculating total error, and doing a backwards pass to update weights.

    Original Description:

    About back propagation

    Original Title

    BackPropagation

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    Backpropagation is a fundamental concept in training neural networks and deep learning models. It is a supervised learning algorithm that updates the weights and biases of a neural network to minimize the error between the predicted and actual outputs. The backpropagation algorithm works by computing the gradient of the loss function with respect to each weight via the chain rule, computing the gradient layer by layer from the last to the first to avoid redundant computation. It involves feeding inputs forward, calculating errors backward, and updating weights. An example shows initializing weights randomly, doing a forward pass, calculating total error, and doing a backwards pass to update weights.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    8 views19 pages

    Back Propagation

    Uploaded by

    Jafer
    Backpropagation is a fundamental concept in training neural networks and deep learning models. It is a supervised learning algorithm that updates the weights and biases of a neural network to minimize the error between the predicted and actual outputs. The backpropagation algorithm works by computing the gradient of the loss function with respect to each weight via the chain rule, computing the gradient layer by layer from the last to the first to avoid redundant computation. It involves feeding inputs forward, calculating errors backward, and updating weights. An example shows initializing weights randomly, doing a forward pass, calculating total error, and doing a backwards pass to update weights.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    of 19

    By

    Akshay Prashad
    Arshath Farwyz
    BACKPROPAGATION Aashiq
    Aashik
    Abdul Bashith
    Tharun
    INTRODUCTION

    Backpropagation, short for "backward propagation of errors," is a fundamental concept in the field of
    artificial neural networks, particularly in the training of deep learning models. It is a supervised learning
    algorithm that plays a crucial role in updating the parameters (weights and biases) of a neural network to
    minimize the error between the predicted output and the actual target output.

    This efficiency makes it possible to use gradient methods to train multi-layer networks and update weights
    to minimize loss; variants such as gradient descent or stochastic gradient descent are often used.

    The backpropagation algorithm works by computing the gradient of the loss function with respect to each
    weight via the chain rule, computing the gradient layer by layer, and iterating backward from the last layer to
    avoid redundant computation of intermediate terms in the chain rule.
    FEATURES OF BACKPROPAGATION

    oIt is the gradient descent method as used in the case of simple perceptron network
    with the differentiable unit.
    oIt is different from other networks in respect to the process by which the weights are
    calculated during the learning period of the network.
    oTraining is done in the three stages :
    I. the feed-forward of input training pattern
    II. the calculation and backpropagation of the error
    III. updation of the weight
    Multi-Layer Neural Network
    BACKPROPAGATION ALGORITHM
    Step 1:
    Inputs X, arrive through the preconnected path.
    Step 2:
    The input is modeled using true weights W. Weights are usually chosen randomly.
    Step 3:
    Calculate the output of each neuron from the input layer to the hidden layer to the
    output layer.
    EXAMPLE
    INITIALIZATION
    FORWARD PASS
    CALCULATING THE TOTAL ERROR
    THE BACKWARDS PASS
    REFERENCES
    https://mattmazur.com/2015/03/17/a-step-by-step-backpropagation-example/
     https://www.geeksforgeeks.org/backpropagation-in-data-mining

    You might also like