lp-Norm Multiple Kernel Learning

Research output: Journal contributionsJournal articlesResearchpeer-review

Authors

Learning linear combinations of multiple kernels is an appealing strategy when the right choice of features is unknown. Previous approaches to multiple kernel learning (MKL) promote sparse kernel combinations to support interpretability and scalability. Unfortunately, this `1-norm MKL is rarely observed to outperform trivial baselines in practical applications. To allow for robust kernel
mixtures that generalize well, we extend MKL to arbitrary norms. We devise new insights on the connection between several existing MKL formulations and develop two efficient interleaved optimization strategies for arbitrary norms, that is `p-norms with p1. This interleaved optimization is much faster than the commonly used wrapper approaches, as demonstrated on several data sets. A theoretical analysis and an experiment on controlled artificial data shed light on the appropriateness of sparse, non-sparse and `¥-norm MKL in various scenarios. Importantly, empirical applications of `p-norm MKL to three real-world problems from computational biology show that non-sparse MKL achieves accuracies that surpass the state-of-the-art.
Original languageGerman
JournalJournal of Machine Learning Research
Volume2011
Issue number12
Pages (from-to)953-997
Number of pages45
ISSN1532-4435
Publication statusPublished - 2011
Externally publishedYes

Recently viewed

Publications

  1. Design optimization of spiral coils for textile applications by genetic algorithm
  2. Exact and approximate inference for annotating graphs with structural SVMs
  3. Fast, Fully Automated Analysis of Voriconazole from Serum by LC-LC-ESI-MS-MS with Parallel Column-Switching Technique
  4. Recurrence Quantification Analysis of Processes and Products of Discourse
  5. Lessons learned for spatial modelling of ecosystem services in support of ecosystem accounting
  6. Construct Objectification and De-Objectification in Organization Theory
  7. Computational modeling of amorphous polymers
  8. Modeling and numerical simulation of multiscale behavior in polycrystals via extended crystal plasticity
  9. Influence of Process Parameters and Die Design on the Microstructure and Texture Development of Direct Extruded Magnesium Flat Products
  10. Simple saturated PID control for fast transient of motion systems
  11. Dynamic Lot Size Optimization with Reinforcement Learning
  12. The delay vector variance method and the recurrence quantification analysis of energy markets
  13. Introducing parametric uncertainty into a nonlinear friction model
  14. Faulty Process Detection Using Machine Learning Techniques
  15. TextGraphs 2024 Shared Task on Text-Graph Representations for Knowledge Graph Question Answering
  16. Clause identification using entropy guided transformation learning
  17. Mathematical Modeling for Robot 3D Laser Scanning in Complete Darkness Environments to Advance Pipeline Inspection
  18. Dispatching rule selection with Gaussian processes
  19. Constraints are the solution, not the problem
  20. Dynamic priority based dispatching of AGVs in flexible job shops
  21. Mining positional data streams
  22. Understanding the properties of isospectral points and pairs in graphs
  23. Improving students’ science text comprehension through metacognitive self-regulation when applying learning strategies
  24. Comments on "Tracking Control of Robotic Manipulators With Uncertain Kinematics and Dynamics"
  25. Computing regression statistics from grouped data
  26. From Knowledge to Application
  27. Gaussian processes for dispatching rule selection in production scheduling