I want to apply neural network as an auto associative memory. So, the desired output is equal to the input. I would apply Hebbs rule to train the network.
I have a pattern in the form
Sample1 = [1 1 1 1 1 1 -1 1 1 1 1 1 -1 -1 -1 -1 -1 -1 1 1 1 1 1 1 1 1 1 1 1 1 ]';
The length d = 30. I have a set of p samples stored in a database,
Database, X = {Sample1,Sample2,....,Sample_p}
But I have some conceptual problem in understanding what determines the input to the neural network -- will it be all samples (example) or each sample /example? In general, what is meant by number of inputs and number of outputs?
答案 0 :(得分：1)
The input to the neural network is one example [1]. The number of inputs is defined as the number of values associated with this example. In your case, the number of inputs is 30.
The number of outputs depends on your task. Since your goal is for the output to be equal to the input, the number of outputs is the same as the number of inputs. Note however, the actual values of the outputs/inputs may not be exactly the same. The number of outputs in your case is also 30.
[1] To speed up the training/learning procedure, machine learning people tend to use more than one example as input simultaneously. This is not always necessary but the fundamental concept is to feed one example at a time.