Which of the following descriptions of the depth feedforward network is correct?

Question

Options

• ADeep feedforward network 1s a kind of neural network.
• BThe deep feedforward network has only one hidden layer.
• CThe unit on the hidden layer of the deep feedforward network will have countless.
• DDeep feedforward network 1s used to deal with linear problems.

• A
  77% (40)
• B
  13% (7)
• C
  6% (3)
• D
  4% (2)

Explanation

A deep feedforward network (also called a multilayer perceptron) is indeed a type of neural network - specifically one where information flows in one direction (input → hidden layers → output) with no feedback loops, making option A correct.

Option B is wrong because "deep" specifically means the network has multiple hidden layers, not just one (a single hidden layer would make it a shallow network). Option C is wrong because hidden layer units are finite and determined by the architecture design - "countless" units would be computationally impossible. Option D is wrong because deep feedforward networks are used precisely to solve non-linear problems; their power comes from non-linear activation functions stacked across layers, which allows them to approximate complex functions that linear models cannot.

Memory tip: Think of the word "deep" - depth implies many layers, ruling out B. And neural networks exist to handle complexity that linear models can't, so D is the opposite of the truth.

No community discussion yet for this question.