In this problem learning rate (l) is fixed as 0.4. We can fix the learning rate to any value in the range 0 to 1 depending on the convergence.. We can also use optimization algorithms to find out optimal learning rate for an application..
Initial weights for Networks and bias will be given in the question ... If the weights are not given then initially we can assign random values and update it at each training iterations during network training process.
Thank you mam.... I'm very much satisfied after watching this video!
Very nice mam... Excellent information mam
Very good explanation
Great madam
mam if we get two updated value can we stop in the first iteration mam ? and test for any input...
Good.
Weight random ah assign pannom ah???
Yes. Only the initial weights
Initial weight yepdi mam potinga?
Initial weights are randomly selected
Mam whether that L = 0.4 value is common for all problem or else how to find out mam
In this problem learning rate (l) is fixed as 0.4. We can fix the learning rate to any value in the range 0 to 1 depending on the convergence.. We can also use optimization algorithms to find out optimal learning rate for an application..
@@Dr.NancyJane OK mam thank you mam
in what range we can use the weights
mam question la w0 kudukalana enna pananum?
Initial weights for Networks and bias will be given in the question ... If the weights are not given then initially we can assign random values and update it at each training iterations during network training process.
Mam kindly put videos on multi layer perceptron also mam....
th-cam.com/video/dOE17M6btYs/w-d-xo.html
Mam kindly send the ppt for this mam