[MXML-11-01] Extreme Gradient Boosting (XGBoost) [1/9] - Regression: Training & Prediction process скачать в хорошем качестве

[MXML-11-01] Extreme Gradient Boosting (XGBoost) [1/9] - Regression: Training & Prediction process

*** Dubbing: [ English ] [ 한국어 ] In the 11th session of Machine Learning, we will discuss the XGBoost called Extreme Gradient Boosting. This video is part 1 of a series on XGBoost. Let's look at the training and the prediction process of XGBoost regression. Let's take a look at the full table of contents. Chapter 1 provides an overview of XGBoost. In Chapters 2 and 3, we will look at the Exact Greedy algorithm. In Chapter 2, we will look at the regression and in Chapter 3, we will look at the classification. The Exact Greedy algorithm is very accurate, but takes a long time when there is a lot of data. In Chapter 4, we will look at the Split Finding algorithm. For large amounts of data, the Exact Greedy algorithm takes a long time, so an approximate method is needed. Let's take a look at the Weighted Quantile Sketch method and also look at parallel processing methods. In Chapter 5, we will look at how to use the xgboost library. In this video, we will look at the training and the prediction process of XGBoost regression. XGBoost is an ensemble learning method proposed by Tianqi Chen and Carlos Guestrin in 2016. This is an improved version of the existing GBM algorithm to effectively process large amounts of data. XGBoost applies the regularization and the pruning to create Decision Trees. And the methods to roughly determine the optimal split points to effectively process large amounts of data were presented. Additionally, various techniques were presented, including missing value handling, parallel processing, and cache-aware access, etc. #ExtremeGradientBoosting #XGBoost