The Statistical Ranking Color Scheme for Visualizing Parameterized Multiple Pairwise Comparisons with R by Pablo J. Villacorta and José A. Sáez is described here.
Steven Ramage from University of British Columbia (UBC) describes the SPO development and the results from the cooperation between UBC and the SPOTSevenLab (Cologne University of Applied Sciences) as follows:
“Bartz-Beielstein et al.  adapted EGO to the optimization of algorithms, with their Sequential Parameter Optimization (SPO) method. SPO uses the same acquisition functions as EGO, but a slightly different model, which includes a second order polynomial fit as well as the standard Gaussian process model. Unlike EGO, their approach is able to deal with random response values through a continual resampling of the best observed points using a doubling strategy, which allows the estimate to converge to the true value over time. Finally, as opposed to fitting the model with each sample point individually, as done by SKO, SPO merges the samples for each point into a better estimate of the objective at that point, and then fits the model on these merged estimates.
Hutter et al. [2009a] directly compared SPO and SKO and their suitability for algorithm configuration. They found that SPO in general outperformed SKO on the algorithms they studied. They also introduced SPO+, which introduced some modifications to the original algorithm….” Continue reading
The new software package for Combinatorial Efficient Global Optimization (CEGO) is now available on CRAN, the distribution network of the free statistical computing language R.
Surrogate-model based optimization algorithms enable to tackle expensive or time-consuming
optimization problems. Such problems often arise from complex simulations or real world experiments. Until recently, surrogate-modelling was rarely considered in the context of combinatorial optimization. CEGO provides several methods for modelling and surrogate-model based optimization in combinatorial or mixed search spaces. In their core, the various modelling methods are based on measures of distance (or dissimilarity) between candidate solutions. Thus, model based approaches that are well established in continuous optimization can be extended to combinatorial problems.
Parallel architectures with node-level accelerators promise significant performance improvements over conventional homogeneous systems. To cope with the increased complexity of programming such systems various pattern-based programming libraries have become available . The integration of pattern-based programming libraries with tuning frameworks such as SPOT  is presented in the paper “Autotuning of Pattern Runtimes for Accelerated Parallel Systems“, which was written by Enes Bajrovic, Siegfried Benker, Jiri Dokulil, and Martin Sandrieser (Research Group Scientic Computing, University of Vienna, Austria). Continue reading
The Korea Herald mentions a fingerprint mobile payment system for Android. Users can download an application called BarTong and use their smartphones like cash in convenient stores. Instead of entering a password, they scan their fingerprint. Continue reading
Last week Jörg and Thomas attended the JMP Workshop “Versuchsplanung und Datenvisualisierung für die Prozessentwicklung” in Cologne, where Louis Valente gave an interesting talk about Visual Data Exploration. Louis Valente is the Global Technical Team Enablement Manager at JMP/SAS.
ORCH is a collection of R packages that enables Big Data analytics using HDFS, Hive, and Oracle Database from a local R environment. Version 2.1.0 is now available. Read more in Oracle’s R blog: https://blogs.oracle.com/r/
Am 19. und 20.6.2013 finden an der FH Köln die Tage der Datenbanken statt. Weitere Infos unter http://goo.gl/5NcLM