EasyDeepLearn

How does SVM adapt to regression (SVR)?

medium

Answer

  • SVR uses an epsilon-insensitive loss: residuals with  y    yhat\mathrm{with}\; \mid y\; - \;y_{\mathrm{hat}}| ≤ epsilon are free (zero penalty), residuals outside are penalized linearly (like L1).
  • Combined with the kernel trick, it fits a flexible non-linear function using only a subset of training points as support vectors.
  • Tune C (violation penalty), epsilon (tube width), and gamma (for RBF).
  • Works well on small/medium tabular problems where linear methods underfit and boosted trees are overkill.
Check yourself — multiple choice
  • SVR uses squared error loss
  • SVR uses an epsilon-insensitive loss and support vectors just like classification SVMs
  • SVR only works with linear kernels
  • SVR cannot handle nonlinearities

SVR = ε-insensitive tube + kernel + SVs. Nonlinear regression via kernels.

#svm#regression

Practise Supervised Learning

215 interview questions in this topic.

Related questions