ML Chapter 02

What type of values does a perceptron output?

• Real numbers
• Binary values (1 or 0)
• Characters
• Complex numbers

What condition causes a perceptron to output 1?

• When the signal is less than theta
• When weights are negative
• When the sum of weighted inputs exceeds the threshold
• When the signal is equal to zero

Which logic gate outputs 1 only when both inputs are 1?

• OR
• NAND
• AND
• XOR

Which logic gate outputs 0 only when both inputs are 1?

• XOR
• NAND
• AND
• OR

Which logic gate outputs 1 if at least one input is 1?

• NAND
• AND
• OR
• XOR

Why can't XOR be modeled with a single-layer perceptron?

• It is a non-binary gate
• It requires recurrent layers
• It needs non-linear separation
• It doesn't use weights

What makes it possible to express XOR using a perceptron?

• Increasing input size
• Using only NAND gates
• Adding hidden layers
• Removing the bias

What is a perceptron unable to model with a single layer?

• OR gate
• NAND gate
• AND gate
• XOR gate

What is the purpose of the bias in a perceptron?

• To amplify the output signal
• To reduce computation time
• To shift the activation threshold
• To connect multiple layers

Which function does a perceptron perform when summing inputs and weights?

• Matrix inversion
• Vector normalization
• Linear combination
• Activation encoding