Material Detail

Limitations of kernel and multiple kernel learning

Limitations of kernel and multiple kernel learning

This video was recorded at 1st International Workshop on Similarity-Based Pattern Analysis and Recognition. Many low Vapnik-Chervonenkis (and hence statistically learnable) classes cannot be represented as linear classes in such a way that they can be learnt with large margin approaches. We review these results and then consider recent bounds for multiple kernel learning that suggest large margin methods may be more general applicable.... Show More


  • User Rating
  • Comments
  • Learning Exercises
  • Bookmark Collections
  • Course ePortfolios
  • Accessibility Info

More about this material


Log in to participate in the discussions or sign up if you are not already a MERLOT member.