Approaching evolutionary robotics through population-based incremental learning

Southey, Finnegan; Karray, Fakhri

doi:10.1109/icsmc.1999.825348

Cited by 14 publications

(12 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Much of the ER work to date used very simple network topologies and restricted weight values [11,40,50,52]. Such restrictions limit the scalability of the methods studied.…”

Section: The Evolutionary Neural Networkmentioning

confidence: 98%

Using direct competition to select for competent controllers in evolutionary robotics

Nelson¹,

Grant²

2006

Robotics and Autonomous Systems

View full text Add to dashboard Cite

“…Much of the ER work to date used very simple network topologies and restricted weight values [11,40,50,52]. Such restrictions limit the scalability of the methods studied.…”

Section: The Evolutionary Neural Networkmentioning

confidence: 98%

Using direct competition to select for competent controllers in evolutionary robotics

Nelson¹,

Grant²

2006

Robotics and Autonomous Systems

View full text Add to dashboard Cite

“…PBIL belongs to the family of Estimation of Distribution Algorithms (EDAs), which use the probability (or prototype) vector (PV) to generate sample solutions. There is no crossover operator in PBIL; instead a single probability vector is updated using solution with the highest fitness values [16]. Initially, the values of the probability vector are set to 0.5 to ensure that the probability of generating 0 or 1 is equal.…”

Section: Overview Of Population-based Incremental Learningmentioning

confidence: 99%

“…PBIL is simpler and more effective than GAs. In addition PBIL has less overhead than GAs [15]- [16]. There are few parameters to tune in PBIL as compared to GAs or PSO.…”

Section: Introductionmentioning

confidence: 99%

Power system controller design using multi-population PBIL

Folly

Venayagamoorthy

2013

2013 IEEE Computational Intelligence Applications in Smart Grid (CIASG)

View full text Add to dashboard Cite

The application of a multi-population based Population-Based Incremental Learning (PBIL) to power system controller design is presented in this paper. PBIL is a combination of evolutionary optimization and competitive learning derived from artificial neural networks. Single population PBIL has recently received increasing attention in various engineering fields due to its effectiveness, easy implementation and robustness. Despite these strengths, PBIL still suffers from issues of loss of diversity in the population. The use of multi-population is seen as one way of increasing the diversity in the population. The approach is applied to power system controller design. Simulations results show that the multipopulation PBIL approach performs better than the standard PBIL and is as effective as PBIL where adaptive learning is used.

show abstract

“…The main benefits of PBIL over traditional evolutionary algorithm approaches are (1) lowered memory requirements (since the population does not needed to be stored); (2) increasing the sample size has no effect on memory (versus increasing population size); and (3) typically a lowered computational cost (mainly because of not employing genetic operators) [22]. Nevertheless, PBIL (and generally EDAs) still suffer from issues of diversity loss, as described in [2,20,36], and so methods which improve diversity should be of importance.…”

Section: Introductionmentioning

confidence: 99%

“…Additionally, for large disturbances in conditions PBIL is shown to outperform a genetic algorithm ap-proach. An initial investigation into the use of PBIL for evolutionary robotics was given in [22]. The proposed Floating Point PBIL is able to handle floating point results and is shown to outperform a traditional genetic algorithm.…”

Section: Introductionmentioning

confidence: 99%