描述Error surface of a linear neuron for a single training case.png
English: On the x-axis is the output, y, of a linear neuron and on the y-axis is the error, E, of the output compared to a target output of zero. The error is calculated by squaring the difference between the actual and target output. This demonstrates that training of a neural network can be thought of as an optimization problem where the correct output involves finding the minima of the parabola.
function[output]=Plot_Error_Surface(x, expected)% Plots the error surface for a single output unit with the given% expected output.holdallfore=expectedy=(e-x).^2;plot(x,y)gridontitle('Error Surface for a Single Training Case');xlabel('y')ylabel('E')endholdoffPlot_Error_Surface(-5:0.5:5,0)
{{subst:Upload marker added by en.wp UW}} {{Information |Description = {{en|On the x-axis is the output, y, of a linear neuron and on the y-axis is the error, E, of the output compared to a target output of zero. The error is calculated by squaring the...