Bagging vs Boosting

BaggingBoosting
1.. Base estimators Runs in ParallelBase estimators Runs in Sequence
2.. Base estimators are Independent of each otherBase estimators are Dependent of each other
3.. Individual Trees are Deep Trees that can overfitIndividual Trees are Stumps i.e. Max-Depth=1
4.. Each Tree have equal importanceSome Trees have more importance or Amount of Say as compared to others
5.. Needs more CPU cores and RAMNeeds less CPU cores (because its sequential) and
less RAM (because of stumps)
6.. Examples – Random Forest, ExtraTrees ClassifierExamples – AdaBoost, GBM, XGBoost

AdaBoostGBM
1.. Adaptive Boosting uses SAMME or SAMME.R AlgorithmGradient Boosting uses Gradient Descent Algorithm for Convergence
2.. Some Trees have more importance or Amount of Say as compared to othersNo mention of importance or Amount of Say at individual tree level
3.. No mention of learning rate hereLearning Rate plays a important role here
4.. Sample weights are adjusted at each iterationSample weights are NOT adjusted at each iteration
5.. Each tree predicts for actual outputEach tree predicts for Residual value except the first one that uses Averaging

Rahul Aggarwal
http://guardiancoder.in

Senior Data Scientist and Gen-AI Engineer #DataScience #AI #RNN #CNN #GenAI #ChatGPT #LLMs

Leave a Reply

Discover more from Rahul Aggarwal's EdTech

Subscribe now to keep reading and get access to the full archive.

Continue reading