Consider below training error and test error observed as we train for a neural network using...

 

Transcribed Image Text:

Consider below training error and test error observed as we train for a neural network using batch gradient descent. Error versus Weight Updates (Example 1) 0.01 Training set error Validation set error 0.009 0.008 0.007 0.006 0.005 0.004 0.003 0.002 S000 10000 15000 20000 Number of weight updates 1) [3 pts] Is there overfitting with the trained model? How do you know? 2) [3 pts) i) If we double the size of the training data, plot the new curves (on the figure together with the old curves) for training error and testing error, respectively. [5 pts] ii) Briefly explain why we may have such new curves. Error Consider below training error and test error observed as we train for a neural network using batch gradient descent. Error versus Weight Updates (Example 1) 0.01 Training set error Validation set error 0.009 0.008 0.007 0.006 0.005 0.004 0.003 0.002 S000 10000 15000 20000 Number of weight updates 1) [3 pts] Is there overfitting with the trained model? How do you know? 2) [3 pts) i) If we double the size of the training data, plot the new curves (on the figure together with the old curves) for training error and testing error, respectively. [5 pts] ii) Briefly explain why we may have such new curves. Error

Answer rating: 100% (QA)

Solution 1 Yes since training error is very less and the validation error i... View the full answer