With only limited labeled data available how might a neural network use case be realized?

With only limited labeled data available how might a neural network use case be realized?
A . by assigning random labels
B . by increasing the depth of the neural network
C . by creating random data
D . by using a customized pre-trained model

View Answer

Answer: D