Resonance Bootstrap & Random Forests
Shuai Huang, Houtao Deng · Data analytics · 2021
Decomposing a computational task into repetitive subtasks needs skillful design. The subtasks should be identical in their mathematical forms, so if they differ from each other they only differ in their specific parametric configurations. A subtask should also be easy to solve, and sometimes there is even closed-form solution. These subtasks may or may not be carried out in a sequential manner, but as a whole, they move things forward—solving a problem. There are multiple perspectives to look at Bootstrap. One perspective that has been well studied in the seminar book is to treat Bootstrap as a simulation of the sampling process. The Bootstrap starts from what authors have forgotten. Not to study an apple as an apple, it studies the process of how an apple is created, the process of apple-ing.