neural networks research group
areas
people
projects
demos
publications
software/data
Enhanced Optimization with Composite Objectives and Novelty Selection (2018)
Hormoz Shahrzad
, Daniel Fink and
Risto Miikkulainen
An important benefit of multi-objective search is that it maintains a diverse population of candidates, which helps in deceptive problems in particular. Not all diversity is useful, however: candidates that optimize only one objective while ignoring others are rarely helpful. This paper proposes a solution: The original objectives are replaced by their linear combinations, thus focusing the search on the most useful tradeoffs between objectives. To compensate for the loss of diversity, this transformation is accompanied by a selection mechanism that favors novelty. In the highly deceptive problem of discovering minimal sorting networks, this approach finds better solutions, and finds them faster and more consistently than standard methods. It is therefore a promising approach to solving deceptive problems through multi-objective optimization.
View:
PDF
Citation:
In
Proceedings of the 2018 Conference on Artificial Life
, Tokyo, Japan, 2018.
Bibtex:
@inproceedings{shahrzad:alife18, title={Enhanced Optimization with Composite Objectives and Novelty Selection}, author={Hormoz Shahrzad and Daniel Fink and Risto Miikkulainen}, booktitle={Proceedings of the 2018 Conference on Artificial Life}, address={Tokyo, Japan}, url="http://nn.cs.utexas.edu/?shahrzad:alife18", year={2018} }
People
Risto Miikkulainen
Faculty
risto [at] cs utexas edu
Hormoz Shahrzad
Masters Student
hormoz [at] cognizant com
Areas of Interest
Evolutionary Computation
Multiobjective Optimization
Artificial Life