MLVU 4.4: Principal Component Analysis @riskone1
MLVU 4.4: Principal Component Analysis  @riskone1
Uploaded February 2021 | Updated November 2026, 1 day ago
In the final video of this lecture, we look at what to do if you have too many features. We only look at one tactic in depth, but it's a doozy: principal component analysis. One of the most magical linear methods available in data science. It won't just help you reduce your data. It you use it right, it'll help you understand your data.

For a more in-depth explanation, see http://peterbloem.nl/blog/pca

slides: mlvu.github.io/lectures/22.Methodology2.annotated.pdf
lecturer: Peter Bloem
MLVU 4.4: Principal Component AnalysisMLVU 11.3: Recurrent neural nets and LSTMsMLVU 4.2: Class imbalance and feature design12 Matrix models: Recommender systems, PCA  and Graph convolutionsMLVU 2.4 Linear ClassificationMLVU 1.5 Generalization in machine learningMLVU 8.4: Expectation-maximization from first principlesMLVU Live Stream10 Tree Models and Ensembles: Decision Trees, Boosting, Bagging, Gradient Boosting (MLVU2018)MLVU Course details 2023MLVU 4.3: NormalizationMLVU 13.4: AlphaGo
MLVU |

MLVU 4.4: Principal Component Analysis

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER