Reference no: EM132716390
Question#1
Suppose you have a single neuron with the following weights:
w1=2, w2=-1
What is the output of the activation function for the input x1=1 and x2=2
o 0.4
o 0.5
o 0.8
Question#2
How does a Neural Network simulate a biological neuron?
1. The neuron receives a single signal which is added to a biased term.
2. The input signals are linearly combined with weights: one weight for each signal.
3. The signals are feed into an activation function and the corresponded outputs are then linearly combined with weights plus a biased term.
Question#3
What is a cost function?
1.The cost function is used during the neural network training to find the optimal weights that will maximize the quality of the model: we want to maximize this function.
2.The cost function is used during the neural network training to find the minimal weights that will maximize this function.
3.The cost function is used during the neural network training to find the optimal weights that will maximize the quality of the model: we want to minimize this function.
Question#4
Suppose you have a dataset with 10 features and a single categorical target variable for a total of 11 columns.
The 10 features are numerical and the target variable can assume the following labels: ('High', 'Medium', 'Low').
Based on this information, which of the following architecture is a good candidate?
1. Input Layer: 10 neurons
Hidden Layer: a single layer of 7 neurons
Output Layer: 3 neurons
2. Input Layer: 10 neurons
Hidden Layer: two layers of 4 neurons each
Output Layer: 3 neurons
3. Input Layer: 10 neurons
Hidden Layer: a single layer of 4 neurons
Output layer: 3 neurons