Learning and Intelligent Optimization

1. Front Matter

   PDF

2. Nested Partitioning for the Minimum Energy Broadcast Problem

   • Sameh Al-Shihabi, Peter Merz, Steffen Wolf

   Pages 1-11

3. An Adaptive Memory-Based Approach Based on Partial Enumeration

   • Enrico Bartolini, Vittorio Maniezzo, Aristide Mingozzi

   Pages 12-24

4. Learning While Optimizing an Unknown Fitness Surface

   • Roberto Battiti, Mauro Brunato, Paolo Campigotto

   Pages 25-40

5. On Effectively Finding Maximal Quasi-cliques in Graphs

   • Mauro Brunato, Holger H. Hoos, Roberto Battiti

   Pages 41-55

6. Improving the Exploration Strategy in Bandit Algorithms

   • Olivier Caelen, Gianluca Bontempi

   Pages 56-68

7. Learning from the Past to Dynamically Improve Search: A Case Study on the MOSP Problem

   • Hadrien Cambazard, Narendra Jussien

   Pages 69-80

8. Image Thresholding Using TRIBES, a Parameter-Free Particle Swarm Optimization Algorithm

   • Yann Cooren, Amir Nakib, Patrick Siarry

   Pages 81-94

9. Explicit and Emergent Cooperation Schemes for Search Algorithms

   • Teodor Gabriel Crainic, Michel Toulouse

   Pages 95-109

10. Multiobjective Landscape Analysis and the Generalized Assignment Problem

    • Deon Garrett, Dipankar Dasgupta

    Pages 110-124

11. Limited-Memory Techniques for Sensor Placement in Water Distribution Networks

    • William E. Hart, Jonathan W. Berry, Erik Boman, Cynthia A. Phillips, Lee Ann Riesen, Jean-Paul Watson

    Pages 125-137

12. A Hybrid Clustering Algorithm Based on Honey Bees Mating Optimization and Greedy Randomized Adaptive Search Procedure

    • Yannis Marinakis, Magdalene Marinaki, Nikolaos Matsatsinis

    Pages 138-152

13. Ant Colony Optimization and the Minimum Spanning Tree Problem

    • Frank Neumann, Carsten Witt

    Pages 153-166

14. A Vector Assignment Approach for the Graph Coloring Problem

    • Takao Ono, Mutsunori Yagiura, Tomio Hirata

    Pages 167-176

15. Rule Extraction from Neural Networks Via Ant Colony Algorithm for Data Mining Applications

    • Lale Özbakır, Adil Baykasoğlu, Sinem Kulluk

    Pages 177-191

16. Tuning Local Search by Average-Reward Reinforcement Learning

    • Steven Prestwich

    Pages 192-205

17. Evolution of Fitness Functions to Improve Heuristic Performance

    • Stephen Remde, Peter Cowling, Keshav Dahal, Nic Colledge

    Pages 206-219

18. A Continuous Characterization of Maximal Cliques in k-Uniform Hypergraphs

    • Samuel Rota Bulò, Marcello Pelillo

    Pages 220-233

19. Hybrid Heuristics for Multi-mode Resource-Constrained Project Scheduling

    • Celso Tchao, Simone L. Martins

    Pages 234-242

20. Back Matter

    PDF

This volume collects the accepted papers presented at the Learning and Intelligent OptimizatioN conference (LION 2007 II) held December 8–12, 2007, in Trento, Italy. The motivation for the meeting is related to the current explosion in the number and variety of heuristic algorithms for hard optimization problems, which raises - merous interesting and challenging issues. Practitioners are confronted with the b- den of selecting the most appropriate method, in many cases through an expensive algorithm configuration and parameter-tuning process, and subject to a steep learning curve. Scientists seek theoretical insights and demand a sound experimental meth- ology for evaluating algorithms and assessing strengths and weaknesses. A necessary prerequisite for this effort is a clear separation between the algorithm and the expe- menter, who, in too many cases, is "in the loop" as a crucial intelligent learning c- ponent. Both issues are related to designing and engineering ways of "learning" about the performance of different techniques, and ways of using memory about algorithm behavior in the past to improve performance in the future. Intelligent learning schemes for mining the knowledge obtained from different runs or during a single run can - prove the algorithm development and design process and simplify the applications of high-performance optimization methods. Combinations of algorithms can further improve the robustness and performance of the individual components provided that sufficient knowledge of the relationship between problem instance characteristics and algorithm performance is obtained.

• Artificial intelligence
• algorithms
• combinatorial optimization
• learning
• machine learning
• optimization
• proving
• algorithm analysis and problem complexity

• Dept. Computer Science, University of Bologna, Bologna, Italy

  Vittorio Maniezzo

• Università degli Studi di Trento, Trento, Italy

  Roberto Battiti

• Discrete Math and Complex Systems Department, Sandia National Laboratories, Albuquerque, USA

  Jean-Paul Watson

Policies and ethics