Consider the problem of training a TLU (threshold logic unit) to do logical disj

Question

Consider the problem of training a TLU (threshold logic unit) to do logical disjunction using the error-correction procedure. Suppose that we start with the initial weights all equal to 0, and learning rate c=1. Find a minimal set of training instances that will correctly train the TLU (threshold logic unit) according to the procedure. Here a training set is minimal if removing any instance in it will not produce a TLU (threshold logic unit) for the logical disjunction. Please show the details of your work including the converging sequence of the weights

Explanation / Answer

In order to train the TLU we will need to have at least 2^n training data where n is the number of inputs to TLU.

This will ensure that the input data ideally tries to represent the truth table. There can be addditional data which can be provided as well which will help fine tune the weights and help rectify any anamolies but these are not mandatory.

e.g, For 3 input TLU minimum number of inputs required will be 2^3 = 8

Related Questions

Navigate

• Browse (All)
• Browse C
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.