Markov Chain methods for the Bipartite Boolean Quadratic Programming Problem
- Submitting institution
-
University of Nottingham, The
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 1328189
- Type
- D - Journal article
- DOI
-
10.1016/j.ejor.2017.01.001
- Title of journal
- European Journal of Operational Research
- Article number
- -
- First page
- 494
- Volume
- 260
- Issue
- 2
- ISSN
- 0377-2217
- Open access status
- Compliant
- Month of publication
- January
- Year of publication
- 2017
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
2
- Research group(s)
-
-
- Citation count
- 11
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The paper provides a novel framework to automate the generation of metaheuristics from available components, and applies it to a specific hard problem, which captures many problems on bipartite graphs. Automated generation of metaheuristics reduces the time needed to build algorithms and improves maintainability. The system produced state-of-the-art results but also, in effect, created metaheuristics that were “counter-intuitive but surprisingly effective”. The work demonstrates that human experts can have too limited a view of design of algorithms, and that automated generation can find better search methods that a human expert may not even consider.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
