PswarmEpochsParallel {DatabionicSwarm}

Intern function, do not use yourself

Description

Finds the weak Nash equilibirium of the data bots for one epoch depending on a radius, which requires the setting of constants, grid, and so on in, see Pswarm.

Usage

PswarmEpochsParallel(AllDataBotsPosRe, AllDataBotsPosIm, MyDistanceMatrix,
AllFreePosR0, GridRadii, GridAngle, JumpsPerRadius, NumJumps, NumAllDB, Lines,
Columns, Origin, Happiness, QuadOrHexa, RadiusVector, Rmin, Rmax, Cls, Debug,
pp, PlotIt = FALSE, Verbose = 1, Eps = 0.0001)

Arguments

AllDataBotsPosRe

Numeric vector [1:n] of the current positions for the databots on first of two dimensions.

AllDataBotsPosIm

Numeric vector [1:n] of the current positions for the databots on second of two dimensions.

MyDistanceMatrix

Numeric vector with vectorized distance matrix of the datapoints in the original (high-dimensional) data space

AllFreePosR0

NumericMatrix, see AllallowedDBPosR0 in setPolarGrid

GridRadii

Numeric matrix with radius information of polar transformation for each grid position

GridAngle

Numeric matrix with angle information of polar transformation for each grid position

JumpsPerRadius

Numeric Vector of possible positions of the 1st coordinate.

NumJumps

Integer number of jumps.

NumAllDB

Integer total number of databots

Lines

Integer stating the number of Lines the polar grid consists of.

Columns

Integer stating the number of columns the polar grid consists of.

Origin

Numeric origin of the positions of grid in two dimensions

Happiness

Numeric value indicating the global happiness over all databots

QuadOrHexa

optional, bool: If TRUE prints status every 100 iterations

RadiusVector

Numeric vector stating all moving radius in a descending order (cooling down scheme).

Rmin

Integer stating minimum radius.

Rmax

Integer stating maximum radius.

Cls

Integer vector stating the classification vector for each datapoints/databots.

Debug

optional, bool: If TRUE prints information for debugging.

pp

Numeric vector stating ratio of number of jumping simultaneously DataBots of one eppoch (per nash-equilibirum), this vector is linearly monotonically decreasing.

PlotIt

optional, bool: If TRUE creates plot of projection after each epoch.

Verbose

optional, integer stating degree of textual feedback. 0 = no output, 1 = basic notifications, 2 = progress bar, 3 = details.

Eps

optional, double: Stop criterion for convergence of each epoche.

Details

Algorithm is described in [Thrun, 2018, p. 95, Listing 8.1].

Value

list of

AllDataBotsPosRe

Numeric vector [1:n] of the current positions for the databots on first of two dimensions.

AllDataBotsPosIm

Numeric vector [1:n] of the current positions for the databots on second of two dimensions.

CourseOfHappiness

NumericVector, states the global happiness value per epoch.

RadiusPerEpoch

NumericVector, stating the radius used per epoch in order of computation.

Author(s)

Quirin Stier

References

[Thrun, 2018] Thrun, M. C.: Projection Based Clustering through Self-Organization and Swarm Intelligence, doctoral dissertation 2017, Springer, Heidelberg, ISBN: 978-3-658-20539-3, doi:10.1007/978-3-658-20540-9, 2018.

[Thrun/Ultsch, 2021] Thrun, M. C., and Ultsch, A.: Swarm Intelligence for Self-Organized Clustering, Artificial Intelligence, Vol. 290, pp. 103237, doi:10.1016/j.artint.2020.103237, 2021.

[Stier/Thrun, 2024] Stier, Q. and Thrun, M. C.: An efficient multicore CPU implementation of the DatabionicSwarm, 18th conference of the International Federation of Classification Societies (IFCS), San José, Costa Rica, July 14-19, 2024.