07 Linear Models 2: Support Vector Machines, the Kernel trick (MLVU2018) @riskone1
07 Linear Models 2: Support Vector Machines, the Kernel trick (MLVU2018)  @riskone1
Uploaded March 2018 | Updated November 2026, 5 days ago
In this lecture, we first review some of the more important details of the linear model, building up our intuition for what it does, and how it works. We then tackle support vector machines in two ways: first as a loss function for general models, and second as a separate classifier, that allows us to apply something called the kernel trick. Lecturer: Peter Bloem. See the PDF for image credits.

Corrections:
- In slide 72, there is a factor of sqrt(2) missing in the last extended feature.

Slides: dropbox.com/s/fo1vq0hcpne9yq4/41.LinearModels2.annotated.pdf?dl=0
07 Linear Models 2: Support Vector Machines, the Kernel trick (MLVU2018)MLVU 1.1 What is machine learning?3 Methodology for evaluation: AUC, Bias and Variance (MLVU2020)4 Methodology 2: Data cleaning, Principal Component Analysis, Eigenfaces (MLVU2019)MLVU 4.1: Missing values and outliers11 Models for Sequential Data: Markov Models, Word2Vec, RNNs and LSTMs.5 Probability 1: Entropy, (Naive) Bayes, Cross-entropy loss (MLVU2019)MLVU 5.2: Learning with probabilityMLVU 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 convolutions
MLVU |

07 Linear Models 2: Support Vector Machines, the Kernel trick (MLVU2018)

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER