Neural Network Foundations

Five questions. Choose one answer for each, then submit to see your score and explanations.

The practice target is 80%. You can retry as often as you like, and every lesson stays accessible.

You can practice as a guest. Sign in before starting an attempt to save your scores.

1. What changes when a neural network learns from examples?A. The saved test answersB. The number of input examples during every predictionC. Its trainable weights and biasesD. The meaning of the task labels

2. Why put nonlinear activation functions between layers?A. They force every output to be a probabilityB. They let the network represent nonlinear relationshipsC. They guarantee perfect test accuracyD. They remove the need for data

3. Which task predicts a continuous numerical value?A. RegressionB. ClassificationC. ClusteringD. Label encoding

4. A neuron computes two times three, plus negative one times four, plus a bias of one. What is its value before activation?A. 1B. 7C. 11D. 3

5. A model memorizes its training pictures. Which check best tests whether it can handle new pictures?A. Add more layers without measuring performanceB. Show it the test labels before predictingC. Evaluate on representative images kept out of trainingD. Evaluate on the same training images only

Review this module from its first lesson →

Neural Network Foundations

Your result