Special Issue on Metalearning and Algorithm Selection

Articles (10 in this collection)

1. Metalearning and Algorithm Selection: progress, state of the art and introduction to the 2018 Special Issue

   Authors

   • Pavel Brazdil
   • Christophe Giraud-Carrier

   • Content type: Editorial
   • Published: 29 December 2017

   • Pages: 1 - 14

2. Instance spaces for machine learning classification

   Authors (first, second and last of 4)

   • Mario A. Muñoz
   • Laura Villanova
   • Kate Smith-Miles

   • Content type: OriginalPaper
   • Published: 28 December 2017

   • Pages: 109 - 147

   

3. Efficient benchmarking of algorithm configurators via model-based surrogates

   Authors (first, second and last of 5)

   • Katharina Eggensperger
   • Marius Lindauer
   • Kevin Leyton-Brown

   • Content type: OriginalPaper
   • Published: 22 December 2017

   • Pages: 15 - 41

   

4. Scalable Gaussian process-based transfer surrogates for hyperparameter optimization

   Authors

   • Martin Wistuba
   • Nicolas Schilling
   • Lars Schmidt-Thieme

   • Content type: OriginalPaper
   • Published: 22 December 2017

   • Pages: 43 - 78

   

5. Meta-QSAR: a large-scale application of meta-learning to drug design and discovery

   Authors (first, second and last of 7)

   • Ivan Olier
   • Noureddin Sadawi
   • Ross D. King

   • Content type: OriginalPaper
   • Open Access
   • Published: 22 December 2017

   • Pages: 285 - 311

   

6. Data complexity meta-features for regression problems

   Authors (first, second and last of 5)

   • Ana C. Lorena
   • Aron I. Maciel
   • Ricardo B. C. Prudêncio

   • Content type: OriginalPaper
   • Published: 21 December 2017

   • Pages: 209 - 246

   

7. Discovering predictive ensembles for transfer learning and meta-learning

   Authors

   • Pavel Kordík
   • Jan Černý
   • Tomáš Frýda

   • Content type: OriginalPaper
   • Published: 21 December 2017

   • Pages: 177 - 207

   

8. The online performance estimation framework: heterogeneous ensemble learning for data streams

   Authors (first, second and last of 4)

   • Jan N. van Rijn
   • Geoffrey Holmes
   • Joaquin Vanschoren

   • Content type: OriginalPaper
   • Open Access
   • Published: 21 December 2017

   • Pages: 149 - 176

   

9. Empirical hardness of finding optimal Bayesian network structures: algorithm selection and runtime prediction

   Authors (first, second and last of 5)

   • Brandon Malone
   • Kustaa Kangas
   • Petri Myllymäki

   • Content type: OriginalPaper
   • Published: 20 December 2017

   • Pages: 247 - 283

   

10. Speeding up algorithm selection using average ranking and active testing by introducing runtime

    Authors (first, second and last of 4)

    • Salisu Mamman Abdulrahman
    • Pavel Brazdil
    • Joaquin Vanschoren

    • Content type: OriginalPaper
    • Published: 14 November 2017

    • Pages: 79 - 108