No Free Lunch Theorems for Search
Explore this paper's citation graph
Summary
It is shown that all algorithms that search for an extremum of a cost function perform exactly the same, when averaged over all possible cost functions, which allows for mathematical benchmarks for assessing a particular search algorithm's performance.
- Type
- preprint
- Published
- 1995-02-01
- Cited by
- 1,278
- References
- 21
- OpenAlex
- https://openalex.org/W1503705371
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:12890367
Keywords
Minimax, A priori and a posteriori, Function (biology), Mathematical optimization, Search algorithm
References
- Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence
- Adaptive Simulated Annealing (ASA)
- Heuristics : intelligent search strategies for computer problem solving
- A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection
- Statistical Decision Theory and Bayesian Analysis, Second Edition
- Neural Network Exploration Using Optimal Experiment Design
- Dynamic Hill Climbing: Overcoming the limitations of optimization techniques
- OFF-TRAINING SET ERROR AND A PRIORI DISTINCTIONS BETWEEN LEARNING ALGORITHMS
- Conservation of Generalization: A Case Study
- Alpha, Evidence, and the Entropic Prior
Cited by
- Methodologies and Software Engineering for Agent Systems
- Fuzzy ARTMAP and a Hybrid Chaos Genetic Algorithm for Medical Pattern Classification
- The road to everywhere : Evolution, complexity and progress in natural and artificial systems
- Back to reality: Crossing the reality gap in evolutionary robotics
- Robust statistical techniques for the categorization of images using associated text
- An Indexed Bibliography of Genetic Algorithms Theory and Comparisons
- Evolutionary methods for musical composition
- Developing Efficient Metaheuristics for Communication Network Problems by using Problem-specific Knowledge
- Towards a Better Forecasting Model for Economic Indices
- Learning to identify new information
- Transfer Learning Using the Minimum Description Length Principle with a Decision Tree Application
- Genetic production systems for intelligent problem solving
- Phlegmatic Mappings for Function Optimization with Genetic Algorithms
- The No Free Lunch and problem description length
- A Serendipitous Software Framework for Facilitating Collaboration in Computational Intelligence
- Comparative evaluation of MAL algorithms in a diverse set of ad hoc team problems
- Attribute Interactions in Machine Learning
- Machine learning in bioinformatics: Algorithms, implementations and applications.
- Large-scale distributed coalition formation
- The Chemnitz Hybrid Evolutionary Optimization System
Related papers
- Genetic Algorithms in Search Optimization and Machine Learning
- No free lunch theorems for optimization
- Optimization by Simulated Annealing
- Genetic Programming: On the Programming of Computers by Means of Natural Selection
- Evolutionsstrategie : Optimierung technischer Systeme nach Prinzipien der biologischen Evolution
- Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence
- Genetic Algorithms
- Particle swarm optimization