MLVU 6.5: The kernel trick @riskone1
MLVU 6.5: The kernel trick  @riskone1
Uploaded February 2021 | Updated November 2026, 2 days ago
ERRATA: The dual problem of the SVM objective says minimize, where it should say _maximize_.

We finish up the lecture by applying what we learned in the previous video to our SVM objective, and rewriting it to a form that contains only the KKT multipliers as parameters. This allows us to use the kernel trick, a powerful idea, that lets us train in high dimensional feature spaces, without explicitly computing lots of features.

slides: mlvu.github.io/lectures/32.LinearModels2.annotated.pdf
lecturer: Peter Bloem
MLVU 6.5: The kernel trick14 Review and outlook (MLVU2018)02 Linear Models 1: Random Search, Gradient Descent (MLVU2018)MLVU 9.3: AutoencodersMLVU Transformers (3/3): Building chatbotsMLVU 13.2: Policy gradients and random searchMLVU 7.1: Deep learning and automatic differentiation3 Methodology 1: Area-under-the-curve, bias and variance, no free lunch (MLVU2019)MLVU 11.1: Markov modelsMLVU 12.3: PCA revisited14 Review: Inductive Bias, Algorithmic Bias, Social impact of machine learning (MLVU2019)MLVU 13.3: (Deep) Q Learning
MLVU |

MLVU 6.5: The kernel trick

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER