XGBoost is a scalable machine learning system for tree boosting Its objective function has a training loss and a regularization term L i l ( ) , y i ) k ( f k ) L i l ( y i , y i ) k ( f k ) Read the XGBoost paper and answer the following questions a What is y i y i At the t t th iteration, XGBoost...

a The final prediction for a given example is the sum of predic ...

XGBoost is a scalable machine learning system for tree boosting. Its objective function has a training loss

Question:

XGBoost is a scalable machine learning system for tree boosting. Its objective function has a training loss and a regularization term: $L = \sum_{i} l (y_{i}, {\hat{y}}_{i}) + \sum_{k} Ω (f_{k})$ . Read the XGBoost paper and answer the following questions:

a. What is ${\hat{y}}_{i}$ ? At the $t$ th iteration, XGBoost fixes $f_{1}, \dots, f_{t - 1}$ and trains the $t$ th tree model $f_{t}$ . How does XGBoost approximate the training $loss l (y_{i}, {\hat{y}}_{i})$ here?