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

Activities

  1. Workshop on "The State and beyond: Actor constellations in resource conflicts" - 2015
  2. 4th Global TraPs Workshop "Defining Case Studies – Setting Priorities”
  3. Workshop of the Nordic Research Network in Memory Studies - 2013
  4. The golden age of software architecture better named the middle age of software architecture - Some provocative thoughts
  5. Enhancing EFL classroom instruction via an ICALL platform: effects on language development and transfer to tasks (EUROCALL)
  6. Corrosion, scaling and biofouling processes in thermal systems and monitoring using redox potential measurements
  7. Learning Processes in a Video-based Learning Environment: What do teachers think and feel when they observe their own teaching or that of others?
  8. Towards an Emotional Geography of Urban Policing: Exploring the Materialization of Police Territoriality with Emotional Mapping Interviews
  9. Micro and macro scale behavior of thermochemical materials in pure and composite forms for thermal storage applications
  10. Linking Teaching and Learning Formats with Student Development of Key Sustainability Competencies
  11. Working in Research-Practice-Partnerships: Empirical Findings on Motivation, Co-Construction and Learning Effects
  12. Keeping drivers engaged in automated driving through maneuver control- effects on perceived control and responsibility
  13. Trajectory-based Lagrangian approaches for the extraction and characterization of coherent structures in turbulent convection
  14. Disaggregating Democracy and the Legitimization of Functionally Fragmented Governance beyond the State
  15. LC-MS identification of the photo-transformation products of desipramine with studying the effect of different environmental variables on the kinetics of their formation
  16. DSP-Kolloquium 2017
  17. From Quantity to Quality: Structuring Provenance Data.
  18. Legibility in the German Deportation Regime: Identification and the Construct of the 'Duldung'

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. Recurrence Quantification Analysis of Processes and Products of Discourse
  4. Lessons learned for spatial modelling of ecosystem services in support of ecosystem accounting
  5. Clause identification using entropy guided transformation learning
  6. Mathematical Modeling for Robot 3D Laser Scanning in Complete Darkness Environments to Advance Pipeline Inspection
  7. An analytical approach to evaluating nonmonotonic functions of fuzzy numbers
  8. An analytical predictor machine learning corrector scheme for modeling lateral flow in hot strip rolling
  9. Improving students’ science text comprehension through metacognitive self-regulation when applying learning strategies
  10. “Ideation is Fine, but Execution is Key”
  11. Comments on "Tracking Control of Robotic Manipulators With Uncertain Kinematics and Dynamics"
  12. From Knowledge to Application
  13. Neural correlates of the enactment effect in the brain
  14. How Much Tracking Is Necessary? - The Learning Curve in Bayesian User Journey Analysis
  15. Data based analysis of order processing strategies to support the positioning between conflicting economic and logistic objectives
  16. Optimization of 3D laser scanning speed by use of combined variable step
  17. Machine Learning and Knowledge Discovery in Databases
  18. Modelling biodegradability based on OECD 301D data for the design of mineralising ionic liquids
  19. Efficient Order Picking Methods in Robotic Mobile Fulfillment Systems
  20. Towards Advanced Learning in Dispatching Rule-Based Scheuling
  21. Learning and Re-learning from net- based cooperative learning discourses
  22. Using Heider’s Epistemology of Thing and Medium for Unpacking the Conception of Documents: Gantt Charts and Boundary Objects
  23. Conceptual understanding of complex components and Nyquist-Shannon sampling theorem
  24. "And I Think That Is a Very Straightforward Way of Dealing With It''
  25. Integrating Common Ground and Informativeness in Pragmatic Word Learning
  26. Dichotomy or continuum? A global review of the interaction between autonomous and planned adaptations
  27. Migration-Based Multilingualism in the English as a Foreign Language Classroom
  28. Requests for mathematical reasoning in textbooks for primary-level students
  29. Context-sensitive adjustment of pointing in great apes
  30. Machine Learning Applications in Convective Turbulence

Press / Media

  1. Wieder gefragt