Biologically Inspired Machine Learning (Perceptrons Contd.)

9 important questions on Biologically Inspired Machine Learning (Perceptrons Contd.)

The    can train a perception by adjusting the  in a single layer.

Perceptron Learning Rule, Weights

   is a set of correctly pre-classified inputs that is used to teach the perceptron to distinguish between two types of inputs.

Training Set

   is another set of correctly pre-classified inputs used after training is complete in order to determine how accurately the system has learned to classify the inputs.

Testing Set
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

The training set and testing set should be .

Disjoint (a single input should not be in both sets).

  occur when the perceptron fires when it should not have.

False Positive

    occur when the perceptron does not fire when it should have.

False Negative

Is the learning process guaranteed to halt?

Yes, if the problem is “perceptron learnable”
No, if the problem is not “perceptron learnable”

What are the characteristics of a “perceptron learnable” problem?

For a two-input problem the problem must be linearly separable.

    means that if you plot the elements of the training set on a two-dimensional grid, you must be able to separate the two output groups by a single straight line.

Linearly Separable

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo