Algorithm portfolio selection as a bandit problem with unbounded losses

DOI10.1007/s10472-011-9228-zzbMath1234.68339MaRDI QIDQ408989

Matteo Gagliolo, Jürgen Schmidhuber

Publication date: 12 April 2012

Published in: Annals of Mathematics and Artificial Intelligence (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/s10472-011-9228-z

zbMATH Keywords

computational complexity; combinatorial optimization; survival analysis; online learning; satisfiability; algorithm portfolios; constraint programming; algorithm selection; multi-armed bandit problem; Las Vegas algorithms; meta learning

Mathematics Subject Classification ID

68Q25: Analysis of algorithms and problem complexity

68T05: Learning and adaptive systems in artificial intelligence

62N99: Survival analysis and censored data

68T20: Problem solving in the context of artificial intelligence (heuristics, search strategies, etc.)

62G99: Nonparametric inference

68W27: Online algorithms; streaming algorithms

Uses Software

• SATzilla
• SPOT