Dunzo Interview Question | Improve Model
Question
You came to know that your model is suffering from low bias and high variance. How will you tackle this situation?
in progress
0
Machine Learning
55 years
3 Answers
861 views
Great Grand Master 0
Answers ( 3 )
Low bias basically means our model is predicting close to actual values. It sounds great but when we do prediction on unseen data then we get poor results. That is when we use Bagging algorithm like Random Forest.
We can use ensemble techniques like bagging, boosting. These algorithms convert the high variance into the low variance. The concept of bagging is that it takes the different sample of the data from the population data and pass it to the different decision tree model to train them and after that it decide the result by the mejority or by taking out the average.
We can use ensemble techniques like bagging, boosting. These algorithms convert the high variance into the low variance.