Author Archives: zaefferer

“Underwater Acoustic Networks for Security Risk Assessment in Public Drinking Water Reservoirs”

Our team here in Gummersbach has recently released an early version of our work on detecting security risks in public drinking water reservoirs. This work is based on different sensors which are suitable to detect objects and sounds under water, and uses neural networks to detect anormal events as well as specific events. Feel free to have a look at this paper, which is available on arXiv (external link):
https://arxiv.org/abs/2107.13977

Interesting talks at GECCO 2021, by the IDEA Team

There have been som interesting talks by my colleagues and co-authors at GECCO 2021:

Thomas Bartz-Beielstein presented a keynote at the SAEOpt Workshop, talking about surrogate model-based tuning of deep learning algorithms:
https://www.spotseven.de/gecco-2021-contributions-from-the-spotseven-lab/

Sowmya Chandrasekaran presented the GECCO industrial challenge:
https://www.spotseven.de/great-presentation-from-sowmya-chandrasekaran-at-gecco-competition-on-optimization-of-a-simulation-model-for-a-capacity-and-resource-planning-task-for-hospitals-under-special-consideration-of-the-c/

Jörg Stork presented our paper on behavior-based neuroevolutionary training in reinforcement learning:
https://www.spotseven.de/jorg-stork-talking-about-behavior-based-neuroevolutionary-training-in-reinforcement-learning-at-gecco-2021/

Updated – on CRAN: CEGO Version 2.4.2

A new version (2.4.2) of my R-package CEGO was recently uploaded to CRAN.

This update contained an important fix to one of the distance functions: distancePermutationSwap. As kindly reported by Manuel López-Ibáñez, that distance was previously computing the swap distance on the inverse of the provided permutations. This is now corrected. The original behavior can be reproduced via distancePermutationSwapInv.

This issue and the impact it can have on the results with CEGO in optimization runs is discussed in an article by Irurozki and López-Ibáñez, which also covers an interesting modeling approach based on Mallows models, see:

Ekhine Irurozki and Manuel López-Ibáñez. Unbalanced Mallows Models for Optimizing Expensive Black-Box Permutation Problems. In Proceedings of the Genetic and Evolutionary Computation Conference, GECCO 2021. ACM Press, New York, NY, 2021. doi: 10.1145/3449639.3459366

For the updated CEGO Package see also (external link) https://cran.r-project.org/package=CEGO.

GECCO 2021 Industrial Challenge

The IDE+A team at TH Köln is hosting the ‘Industrial challenge’ at the GECCO 2021. The goal is to find optimal parameters for the BaBSim.Hospital simulator. Participants can publish their algorithms and the top 3 solutions will be honored with price money.

More Information: 

https://www.th-koeln.de/informatik-und-ingenieurwissenschaften/gecco-2021-industrial-challenge-call-for-participation_82086.php

Example code for talk at GECCO 2017: Simulation-based Test Functions for Optimization Algorithms

You can find the corresponding article in the publications section of this website, and the slides in the presentation section.

The R-code below can be used to reproduce the simple example illustrated in the slides and in the article.


# Introductory comments:
#
# To run this script, you need to install R, the language for statistical computing.
# https://cran.r-project.org/
# For ease of use, you may consider RStudio IDE, but this is not required.
# https://www.rstudio.com/
# For a tutorial/introduction to R, see e.g.:
# https://cran.r-project.org/doc/manuals/r-release/R-intro.html

# Note: the following line loads the package CEGO
# If it is not installed, please do so first, using: 
# install.packages("CEGO")
# Please note, that you need the most recent version (2.2.0) 
# of the package CEGO.
require(CEGO)

nsim <- 10 # number of simulations / realizations
seed <- 12121212 # RNG seed
set.seed(seed)
m <- 100 # number of samples to be simulated at

# objective  function:
fun <- function(x){
  exp(-20* x) + sin(6*x^2) + x
}
# "vectorize" target
f <- function(x){sapply(x,fun)}

# distance function (for model/kernel)
dF <- function(x,y)(sum((x-y)^2))

# plot parameters
par(mfrow=c(3,1),mar=c(2.3,2.5,0.2,0.2),mgp=c(1.4,0.5,0))

# create test samples for plots
xtest <- as.list(seq(from=-0,by=0.005,to=1))
plot(xtest,f(xtest),type="l",lty=2,xlab="x",
				ylab="Estimation",ylim=c(-0.5,1.7))

# evaluation samples (training data)
xb <- c(0.1,0.4,0.54,0.6,0.8,0.99)
yb <- f(xb)

# create support samples for simulation
x <- as.list(sort(c(runif(m-length(xb)),unlist(xb))))

# fit the model	
fit <- modelKriging(xb,yb,
		control=list(distanceFunction=dF,
                algThetaControl=
                list(method="NLOPT_GN_DIRECT_L",
                     funEvals=100),useLambda=F))
fit

# predicted obj. function values
ypred <- predict(fit,as.list(xtest))$y
lines(unlist(xtest),ypred)
points(unlist(xb),yb,pch=19)

##############################	
# create test functions with non conditional sim.
##############################

fun <- createSimulatedTestFunction(x,fit,nsim,F,seed=1)

ynew <- NULL
for(i in 1:nsim)
  ynew <- cbind(ynew,fun[[i]](xtest))

rangeY <- range(ynew)

plot(unlist(xtest),ynew[,1],type="l",
     xlab="x",ylab="Non-conditional",ylim=c(-0.5,1.7))
for(i in 2:nsim){
  lines(unlist(xtest),ynew[,i],col=i,type="l")
}

##############################	
# create test functions with conditional sim.
##############################

fun <- createSimulatedTestFunction(x,fit,nsim,T,seed=1)

ynew <- NULL
for(i in 1:nsim)
  ynew <- cbind(ynew,fun[[i]](xtest))

rangeY <- range(ynew)

plot(unlist(xtest),ynew[,1],type="l",
     xlab="x",ylab="Conditional",ylim=c(-0.5,1.7))
for(i in 2:nsim){
  lines(unlist(xtest),ynew[,i],col=i,type="l")
}
points(unlist(xb),yb,pch=19)



Updated – on CRAN: CEGO Version 2.2.0

A new version (2.2.0) of my R-package CEGO was just uploaded to CRAN.

This update contained some various fixes to code and documentation. Also, the interfacing of C code was reworked, following recent changes to CRAN checks (registering of entry points to compiled code). This may have yielded a speed up for some of the distance calculations. The update also contains code that has been employed in the article Simulation-based Test Functions for Optimization Algorithms (see Publications for the pre-print PDF). This mostly includes functions for Kriging simulation as well as for the generation of corresponding test functions.

See also (external link) https://cran.r-project.org/package=CEGO.

Presentation at 15th Workshop on Quality Improvement Methods

Recently, I had the opportunity to give a talk on the topic of the optimization of the feed material of a biogas plant, at the 15th Workshop on Quality Improvement Methods in Dortmund, Germany. The slides can be found in the presentations section of this site.

I really liked the active discussions at the Workshop. So thanks again to the organizers for inviting me. See also the nice photo of the participants here:

(external link:) https://www.statistik.tu-dortmund.de/qim15.html

You can also find the program and abstracts of all the talks via the above link.

Article Update: Model-based methods for continuous and discrete global optimization

Just a small update for the article mentioned in the last post: Model-based methods for continuous and discrete global optimization.

Until April 11, 2017, the following link will direct you to the final version of the article on ScienceDirect (free, without personal or institutional registration).

(external link:) https://authors.elsevier.com/a/1Ub295aecSVmv2

Recent Publication: Model-based Methods for Continuous and Discrete Global Optimization

The accepted manuscript of the article Model-based Methods for Continuous and Discrete Global Optimization by Thomas Bartz-Beielstein and myself has been published in the Elsevier Journal Applied Soft Computing.

External link: http://dx.doi.org/10.1016/j.asoc.2017.01.039

Among other important issues of Model-based Optimization, the article deals with methods for discrete or combinatorial optimization problems. A tabular survey of works in this field is part of the supplementary material. A PDF of this table will be kept up-to-date on this website, see:

Survey: Combinatorial Surrogate Models

If you have comments or additions with respect to that collection, please let me know.

Also, you can find an earlier preprint version in the publications section.