Homework Help is Here – Start Your Trial Now!
Engineering
AI and Machine Learning
Suppose we are handling a classification problem. The training misclassification error is defined as err(h₂(x(¹)), y(i)), and the cross validation misclassification error is similarly defined by using the cross validation examples (x,y),..., x(mcv), y(mcv)). Suppose the training error is 0.0001, and the cross validation error is 0.40. What problem is the algorithm most likely to be suffering from? Q4 m Li=1 m

Suppose we are handling a classification problem. The training misclassification error is defined as err(h₂(x(¹)), y(i)), and the cross validation misclassification error is similarly defined by using the cross validation examples (x,y),..., x(mcv), y(mcv)). Suppose the training error is 0.0001, and the cross validation error is 0.40. What problem is the algorithm most likely to be suffering from? Q4 m Li=1 m

icon
Related questions
Question

Help to explain the question in following.
Q4)
a.High bias (underfitting
b.High bias (overfitting

c.High variance (underfitting
d.High variance (overfitting


Q5)
a.No,it is currently suffering from high bias, and adding more hidden layers is unlikely to help.
b.No,it is currently suffering from high variance, and adding more hidden layers is unlikely to help.

c.Yes,this increases the number of parameters and lets the network represent more complex functions.
d.Yes,it is currently suffering from high bias.
Suppose we are fitting a neural network with three hidden layers to a training set. It is found that the cross validation error Jcv() is much larger than the training error Jtrain (0). Should we increase the number of hidden layers? Q5
Transcribed Image Text:Suppose we are fitting a neural network with three hidden layers to a training set. It is found that the cross validation error Jcv() is much larger than the training error Jtrain (0). Should we increase the number of hidden layers? Q5
Suppose we are handling a classification problem. The training misclassification error m is defined as err(hö(x(i)),y(i)), and the cross validation misclassification error is similarly defined by using the cross validation examples (x,y),..., m i=1 (mcv), y(mcv)). Suppose the training error is 0.0001, and the cross validation error is 0.40. What problem is the algorithm most likely to be suffering from? 4
Transcribed Image Text:Suppose we are handling a classification problem. The training misclassification error m is defined as err(hö(x(i)),y(i)), and the cross validation misclassification error is similarly defined by using the cross validation examples (x,y),..., m i=1 (mcv), y(mcv)). Suppose the training error is 0.0001, and the cross validation error is 0.40. What problem is the algorithm most likely to be suffering from? 4
Expert Solution
steps

Step by step

Solved in 3 steps

SEE SOLUTIONCheck out a sample Q&A here
Blurred answer