Ensemble Learning Flashcards
Ensemble Learning
Aggregating a group of predictors, you will often get better predictions than with the best individual predictor.
2 Way to Constructing Ensemble Method
By manipulating training set - Example: bagging, boosting, random forests
By manipulating input features - Example: random forests
4 Approach to Manipulating Training Set
Bagging - sampling with replacement
Pasting - sampling without replacement
Random Forests
Boosting: An iterative procedure to adaptively change distribution of training data by focusing more on previously misclassified records
3 Term of Bias-Variance Decomposition
Variance - The distance intervals of predicted value
Noise
Bias - The distance between predicted value and actual value.
Bias-Variance Trade-off
Overfitting - Low Bias & High Variance
Optimum - Low Bias & Low Variance
Underfitting - High Bias & Low Variance
2 Type of Voting Classifiers
Hard Voting Classifiers (Majority-Vote)
Soft Voting Classifiers - Predict the class with the highest class probability, averaged over all the individual classifiers.