Take the 2-minute tour ×
Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It's 100% free, no registration required.

Does it make any sense to have two (or more) neurons in a neural network with the same weights? (intuitively it makes no sense, since all the neurons would behave the same way).

Please consider both the input and hidden layers. What if the weights are equal 0?

share|improve this question
1  
If this is homework, and it really sounds like it from the "please consider" remark, then you should do it yourself, or failing that, say exactly what you have thought about and use the homework tag as stated in the FAQ. –  Douglas Zare Jan 12 '13 at 23:43
add comment

1 Answer

up vote 2 down vote accepted

The weights are normally updated and can have many different values. You can have 2 or more weights with the same value.

If some weights are equal to zero, it just means that the neuron has no impact on the neuron of the next layer.

And one more thing to know is that you shouldn't initialize the values of all your weights at the beginning with the same value. See my previous answer here: stats.stackexchange.com/questions/45087/backpropagation/45092

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.