NOTICE

History of the Geneva library collection
----------------------------------------

The Geneva library collection was initially developed by Dr. Ruediger Berlich
with kind support by the Helmholtz Society and by Steinbuch Centre
for Computing at the Karlsruhe Institute of Technology (as well as
other departments). It was later maintained together with Dr. Ariel Garcia under
the auspices of Gemfony scientific. For further information on Gemfony scientific,
see http://www.gemfomy.eu .

Predecessors of Geneva were developed as of 1994 in the context of a diploma
thesis by Ruediger Berlich at the Institute for Experimental Physics I at
Ruhr-Universität Bochum in order to train feed forward neural networks for the
recognition of hadronic splitoffs at the Crystal Barrel experiment at CERN/Geneva.
That code was extended from 2001 to 2004 under the name "EVA" to support the
optimization of particle physics analysis at the BaBar experiment (SLAC), again at
the Institute for Experimental Physics I at RUB.

Geneva is a complete re-implementation of EVA, with advanced support for distributed
processing of objective functions, added optimization algorithms and a set of
supporting libraries e.g. for acyclic creation of random numbers. It is predominantly used
in physics, but also has a deployment in the automotive industry.

The majority of files of Geneva was released under the Apache v2 license in February 2020
in order to facilitate external contributions to the code and broaden the set of
deployment scenarios. You should have received a copy of that license along with the
Geneva library (see the file licenses/ApacheLicense-2.0). If not, see
<http://www.apache.org/licenses/LICENSE-2.0.txt>.

A small number of files of the Geneva library collection is covered by the Boost
Software License v1. You should have received a copy of the Boost Software License v1
along with the Geneva library (see the file licenses/BoostSoftwareLicense-1.0).



Credits
-------

We'd like to sincerely thank Dr. Sven Gabriel for his kind
support and for discovering numerous problems.

Alexander Müller of the Friedrich-Alexander-Universität Erlangen-
Nürnberg has made many suggestions, discovered bugs and has given
important advice for porting Geneva to Windows.

Lisa Schätzle has helped with the improvement of the Gradient Descent
implementation.

Mathias Michel of Mainz University has kindly shared his experiences
during the deployment of Geneva in a package used for Dalitz plot
analysis and thus helped to mature Geneva.

Dominik Rödel and Felix Zürker of Tenneco Clean Air Europe made
many valuable suggestions during a joint project dealing with the
optimization of the acoustics of exhaust systems.

The work with Prof. Dr. Matthias Lutz and Dr. Kilian Schwarz of
GSI Darmstadt on scalability issues and optimization efficiency
of a particular use case from particle physics has helped to make
Geneva a more mature and versatile solution.

Jonas Wessner has contributed a new MPI consumer that greatly improves
efficiency in a cluster environment, and has analysed many aspects
of Geneva for his Bachelor thesis. His work significantly contributes
to the further development of Geneva.