Aiphabet

Deep Learning Quiz

Need a hint?

Think about how complex patterns are broken down into simpler components, and how this relates to the way information flows through a neural network.

  1. What is the main advantage of having multiple layers in a neural network?

Need a hint?

Consider which activation function produces outputs in the range [0,1] that can be interpreted as a probability of belonging to the positive class.

  1. If you're building a neural network for a binary classification problem (where you need to predict the probability of an outcome belonging to one of two classes), which activation function would be most appropriate for the output layer?

Need a hint?

Remember the three-step process: 1) Calculate the weighted sum of inputs, 2) Add the bias, 3) Apply the activation function. Pay special attention to the sign when dealing with the negative values.

  1. A neuron has inputs [2,1,3][2, -1, 3], weights [0.5,0.5,0.2][0.5, -0.5, 0.2], and bias 11. If it uses a ReLU activation function, what is its output (we call this the neuron's activation)?

Need a hint?

Think about how neural networks "learn" from their mistakes and what mathematical process allows them to improve over time.

  1. What is the purpose of backpropagation in neural networks?

Need a hint?

Feel free to review the table in the first article: Introduction to Deep Learning.

  1. Which are correct matches between the layer type and its primary function in a neural network (select all that apply)?

Need a hint?

Feel free to review the table in the first article: Introduction to Deep Learning.

  1. What is the correct order of steps (in one epoch) in training a neural network?