Strongly Typed Genetic Programming

Montaña, David

doi:10.1162/evco.1995.3.2.199

Cited by 659 publications

(373 citation statements)

References 1 publication

Supporting

Mentioning

371

Contrasting

Unclassified

Order By: Relevance

“…These experiments demonstrate that different tradeoffs can be made between classification accuracy and energy consumption of the programs, and encourage us to find the acceptable trade-offs between these objectives using multi-objective optimisation. Furthermore, since the size of the programs can affect their energy consumption, we conduct experiments to evolve programs with different tree depths (17,5). Tree depth defines the maximum size of the individuals (trees) evolved in GP.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Power-Aware Intrusion Detection in Mobile Ad Hoc Networks

2010

View full text Add to dashboard Cite

Abstract. Mobile ad hoc networks (MANETs) are a highly promising new form of networking. However they are more vulnerable to attacks than wired networks. In addition, conventional intrusion detection systems (IDS) are ineffective and inefficient for highly dynamic and resourceconstrained environments. Achieving an effective operational MANET requires tradeoffs to be made between functional and non-functional criteria. In this paper we show how Genetic Programming (GP) together with a Multi-Objective Evolutionary Algorithm (MOEA) can be used to synthesise intrusion detection programs that make optimal tradeoffs between security criteria and the power they consume.

show abstract

Section: Resultsmentioning

confidence: 99%

“…Here we use strongly-typed GP, which enforces data type constraints and whose use of genetic functions and generic data types [17]. In order to evaluate an individual we translate the individual tree to a C program.…”

Section: F Itness = Detection Rate − F Alse Positive Ratementioning

confidence: 99%

Power-Aware Intrusion Detection in Mobile Ad Hoc Networks

2010

View full text Add to dashboard Cite

show abstract

“…The first method is to make every function of the function set to execute different actions for different data types received as input. The second method is to have these functions returning an error flag when the data types of the received arguments are incompatible, and then assigning an extremely bad fitness value to the corresponding tree [19].…”

Section: Constrained Syntactic Structuresmentioning

confidence: 99%

“…The central issue for syntactically-constrained structures is the definition of the syntactic rules that specify, for each non-terminal, which kinds of child node it can have [15,19].…”

Section: Constrained Syntactic Structuresmentioning

confidence: 99%

A constrained-syntax genetic programming system for discovering classification rules: application to medical data sets

Bojarczuk

Lopes

Freitas

et al. 2004

Artificial Intelligence in Medicine

107

View full text Add to dashboard Cite

This paper proposes a new constrained-syntax genetic programming (GP) algorithm for discovering classification rules in medical data sets. The proposed GP contains several syntactic constraints to be enforced by the system using a disjunctive normal form representation, so that individuals represent valid rule sets that are easy to interpret. The GP is compared with C4.5, a very well-known decision-tree-building algorithm, and with another GP that uses boolean inputs (BGP), in five medical data sets: Chest pain, Ljubljana breast cancer, Dermatology, Wisconsin breast cancer, and Pediatric Adrenocortical Tumor. For this last data set a new preprocessing step was devised for survival prediction. Computational experiments show that, overall, the GP algorithm obtained good results with respect to predictive accuracy and rule comprehensibility, by comparison with C4.5 and BGP.

show abstract

“…We ourselves have used those methods in this article (HAMLET seed, the knowledge-based crossover and the join operator). Also, in GP it is possible to use constrains (to constraint the evolving structures [9,12]) and therefore, they reduce the search space. And of course, it is always possible to select the function and terminals most appropriate to the domain or to the problem.…”

mentioning

confidence: 99%

Evolving heuristics for planning

Aler

Borrafo

Isasi

1998

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In this paper we describe EvoCK, a new approach to the application of genetic programming (GP) to planning. This approach starts with a traditional AI planner (PRODIGY) and uses GP to acquire control rules to improve its efficiency. We also analyze two ways to introduce domain knowledge acquired by another method (HAMLET) into EvoCK:seeding the initial population and using a new operator (knowledge-based crossover). This operator combines genetic material from both an evolving population and a non-evolving population containing background knowledge. We tested these ideas in the blocksworld domain and obtained excellent results. IntroductionProblem solving aims to achieve a set of goals from an initial state by using operators that represent the different actions available in a task domain. Traditional approaches [16,14] use domain independent planners for generating plans. Some more innovative approaches to problem solving use genetic programming [9]. This approach was started by Koza [8,9], who evolved a planner that solved a very specific set of problems in the blocksworld domain. Handley [5][6][7] used GP to evolve plans for specific problems in the blocksworld domain. 1 Muslea [13] generalized, extended, and formalized this idea, and showed how any planning problem could be translated to an equivalent GP problem. He tested it successfully in several domains. Spector [15] proposed and analyzed three ways in which GP could be used for planning: -to evolve a plan for a specific problem in a specific domain (Handley's and Muslea's approach); -to evolve a partially universal planner in a specific domain (like Koza's approach), that is, a planner that is able to solve problems which have the same initial state but different goals states; and -to evolve a fully universal planner in a specific domain, "capable of achieving a range of goal conditions from a range of initial conditions" [15].1 A similar approach in the GA field was used by Davidor [2] to evolve trajectories for a robot arm.

show abstract

Strongly Typed Genetic Programming

Cited by 659 publications

References 1 publication

Power-Aware Intrusion Detection in Mobile Ad Hoc Networks

Power-Aware Intrusion Detection in Mobile Ad Hoc Networks

A constrained-syntax genetic programming system for discovering classification rules: application to medical data sets

Evolving heuristics for planning

Contact Info

Product

Resources

About