Principles of artificial intelligence
Principles of artificial intelligence
Multiple stack branch and bound
Information Processing Letters
Journal of the ACM (JACM)
High-level synthesis: introduction to chip and system design
High-level synthesis: introduction to chip and system design
Multiobjective heuristic search in AND/OR graphs
Journal of Algorithms
On the approximability of trade-offs and optimal access of Web sources
FOCS '00 Proceedings of the 41st Annual Symposium on Foundations of Computer Science
On the approximate tradeoff for bicriteria batching and parallel machine scheduling problems
Theoretical Computer Science
Approximating Multiobjective Knapsack Problems
Management Science
Anytime search in dynamic graphs
Artificial Intelligence
A Multiobjective Branch-and-Bound Framework: Application to the Biobjective Spanning Tree Problem
INFORMS Journal on Computing
Theory of Computing Systems
Near Admissible Algorithms for Multiobjective Search
Proceedings of the 2008 conference on ECAI 2008: 18th European Conference on Artificial Intelligence
Journal of Artificial Intelligence Research
AWA*-a window constrained anytime heuristic search algorithm
IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence
A multiobjective frontier search algorithm
IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence
Limited discrepancy beam search
IJCAI'05 Proceedings of the 19th international joint conference on Artificial intelligence
Multiobjective A* search with consistent heuristics
Journal of the ACM (JACM)
A population and interval constraint propagation algorithm
EMO'03 Proceedings of the 2nd international conference on Evolutionary multi-criterion optimization
A fast and elitist multiobjective genetic algorithm: NSGA-II
IEEE Transactions on Evolutionary Computation
$hbox{MAWA}^{ast}$—A Memory-Bounded Anytime Heuristic-Search Algorithm
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
| Hi-index | 0.00 |
Heuristic search algorithms (MOA*, NAMOA* etc) for multiobjective optimization problems, when run with admissible heuristics, typically spend a lot of time and require huge amount of memory for generating the Pareto optimal solution frontier due to the fact that these algorithms expand all nodes/paths whose cost is not dominated by any other optimal solution. In this paper, we present an anytime heuristic search framework for biobjective optimization problems named "Anytime Biobjective Search (ABS)" which quickly finds a set of nondominated solutions and keeps on improving the solution frontier with time. The proposed framework uses the upper and lower limit estimates on one of the objectives to split the search space into a given number of segments and independently runs a particular search algorithm (branch-and-bound, beam search etc.) within each of the segments. In this paper, we present how existing search strategies, branch-and-bound, beam, and beam-stack, can be used within the proposed framework. Experimental results reveal that our proposed framework achieves good anytime performance.