search

schrum2 / MM-NEAT

Modular Multiobjective (Hyper) Neuro-Evolution of Augmenting Topologies + MAP-Elites: Java code for evolving intelligent agents in Ms. Pac-Man, Tetris, and more, as well as code for Procedural Content Generation in Mario, Zelda, Minecraft, and more!
http://people.southwestern.edu/~schrum2/re/mm-neat.php
Other
50 stars 20 forks source link

Hypervolume calculation and minimum score #944

Closed schrum2 closed 1 year ago

schrum2 commented 1 year ago

In some domains, it is common for individuals to have the minimum score in one objective. This means that the hypervolume calculation would not recognize any improvement in the other objective, since the overall volume would still be 0. Should there be a bit of a buffer around the min score to assure non-zero hypervolume results when just one of several objectives is 0?

schrum2 commented 1 year ago

Introduced hypervolumeMinimumOffset parameter to address this