APPLYING MACHINE LEARNING TO LOW‐KNOWLEDGE CONTROL OF OPTIMIZATION ALGORITHMS
Abstract
This paper addresses the question of allocating computational resources among a set of algorithms to achieve the best performance on scheduling problems. Our primary motivation in addressing this problem is to reduce the expertise needed to apply optimization technology. Therefore, we investigate algorithm control techniques that make decisions based only on observations of the improvement in solution quality achieved by each algorithm. We call our approach “low knowledge” since it does not rely on complex prediction models, either of the problem domain or of algorithm behavior. We show that a low‐knowledge approach results in a system that achieves significantly better performance than all of the pure algorithms without requiring additional human expertise. Furthermore the low‐knowledge approach achieves performance equivalent to a perfect high‐knowledge classification approach.
Citing Literature
Number of times cited: 21
- Eugene C. Freuder, Progress towards the Holy Grail, Constraints, (2017).
- Young Hwan Choi, Donghwi Jung, Ho Min Lee, Do Guen Yoo and Joong Hoon Kim, Improving the Quality of Pareto Optimal Solutions in Water Distribution Network Design, Journal of Water Resources Planning and Management, 143, 8, (04017036), (2017).
- Tiantian Zhang, Michael Georgiopoulos and Georgios C. Anagnostopoulos, Multi-Objective Model Selection via Racing, IEEE Transactions on Cybernetics, 46, 8, (1863), (2016). 2016 IEEE Congress on Evolutionary Computation (CEC) Vancouver, BC, Canada 2016 IEEE Congress on Evolutionary Computation (CEC) IEEE , (2016). 978-1-5090-0623-6 Mario A. Munoz and Michael Kirley ICARUS: Identification of complementary algorithms by uncovered sets , (2016). 2427 2432 7744089 , 10.1109/CEC.2016.7744089 http://ieeexplore.ieee.org/document/7744089/
- Lars Kotthoff, Algorithm Selection for Combinatorial Search Problems: A Survey, Data Mining and Constraint Programming, 10.1007/978-3-319-50137-6_7, (149-190), (2016).
- Roberto Amadini, Maurizio Gabbrielli and Jacopo Mauro, Portfolio approaches for constraint optimization problems, Annals of Mathematics and Artificial Intelligence, 76, 1-2, (229), (2016).
- Ashiqur R. KhudaBukhsh, Lin Xu, Holger H. Hoos and Kevin Leyton-Brown, SATenstein: Automatically building local search SAT solvers from components, Artificial Intelligence, 232, (20), (2016). 16th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference Dallas, TX 16th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference American Institute of Aeronautics and Astronautics Reston, Virginia , (2015). , (2015). 978-1-62410-368-1 10.2514/MMAO15 , 10.2514/MMAO15 2016101707013800537 http://arc.aiaa.org/doi/book/10.2514/MMAO15 Maurice Hoogreef and Gianfranco La Rocca An MDO advisory system supported by knowledge-based technologies , (2015). , (2015). 10.2514/6.2015-2945 , 10.2514/6.2015-2945 2016101707013800537 http://arc.aiaa.org/doi/10.2514/6.2015-2945
- David Corne, Clarisse Dhaenens and Laetitia Jourdan, Synergies between operations research and data mining: The emerging use of multi-objective approaches, European Journal of Operational Research, 221, 3, (469), (2012).
- J. Christopher Beck, T. K. Feng and Jean-Paul Watson, Combining Constraint Programming and Local Search for Job-Shop Scheduling, INFORMS Journal on Computing, 23, 1, (1), (2011). 2010 22nd International Conference on Tools with Artificial Intelligence (ICTAI 2010) Arras 2010 22nd IEEE International Conference on Tools with Artificial Intelligence IEEE , (2010). 978-1-4244-8817-9 Alejandro Arbelaez, Youssef Hamadi and Michele Sebag Continuous Search in Constraint Programming , (2010). 53 60 5670020 , 10.1109/ICTAI.2010.17 http://ieeexplore.ieee.org/document/5670020/
- Roman Barták, Miguel A. Salido and Francesca Rossi, New trends in constraint satisfaction, planning, and scheduling: a survey, The Knowledge Engineering Review, 25, 03, (249), (2010).
- Mario Graff and Riccardo Poli, Practical performance models of algorithms in evolutionary program induction and other domains, Artificial Intelligence, 174, 15, (1254), (2010).
- Mukul Tripathi and Hung-da Wan, Taguchi Integrated Real-Time Optimization for Product Platform Planning, Systems Engineering Tools and Methods, 10.1201/b10452-10, (235-263), (2010).
- Roman Barták, James Little, Oscar Manzano and Con Sheahan, From enterprise models to scheduling models: bridging the gap, Journal of Intelligent Manufacturing, 21, 1, (121), (2010).
- Michela Milano and Alessio Guerri, Bid evaluation in combinatorial auctions: optimization and learning, Software: Practice and Experience, 39, 13, (1127), (2009).
- Tom Carchrae and J. Christopher Beck, Principles for the Design of Large Neighborhood Search, Journal of Mathematical Modelling and Algorithms, 8, 3, (245), (2009).
- HUAYUE WU and PETER VAN BEEK, PORTFOLIOS WITH DEADLINES FOR BACKTRACKING SEARCH, International Journal on Artificial Intelligence Tools, 17, 05, (835), (2008).
- Matteo Gagliolo and Jürgen Schmidhuber, Learning dynamic algorithm portfolios, Annals of Mathematics and Artificial Intelligence, 47, 3-4, (295), (2007). 2007 IEEE Symposium on Computational Intelligence in Scheduling Honolulu, HI, USA 2007 IEEE Symposium on Computational Intelligence in Scheduling IEEE , (2007). 1-4244-0704-4 Ruibin Bai, Edmund K. Burke, Michel Gendreau, Graham Kendall and Barry McCollum Memory Length in Hyper-heuristics: An Empirical Study , (2007). 173 178 4218613 , 10.1109/SCIS.2007.367686 http://ieeexplore.ieee.org/document/4218613/ 19th IEEE International Conference on Tools with Artificial Intelligence(ICTAI 2007) Patras, Greece 19th IEEE International Conference on Tools with Artificial Intelligence(ICTAI 2007) IEEE , (2007). 0-7695-3015-X 978-0-7695-3015-4 Huayue Wu and Peter van Beek On Portfolios for Backtracking Search in the Presence of Deadlines , (2007). 231 238 4410288 , 10.1109/ICTAI.2007.38 http://ieeexplore.ieee.org/document/4410288/
