Quality diversity

tags

Evolution, Reinforcement learning, Search algorithms

papers

(Pugh et al. 2016; Cully, Demiris 2017)

QD is about creating algorithms that favor diversity in searching the space. In QD, one needs to both:

• Measure the quality of a solution
• Have a way to describe the effect of a solution

Solutions in QD have to be good in the two above ways.

QD is also a form of novelty search.

Bibliography

1. Justin K. Pugh, Lisa B. Soros, Kenneth O. Stanley. 2016. "Quality diversity: A new frontier for evolutionary computation". Frontiers in robotics and AI 3. Frontiers. DOI.
2. Antoine Cully, Yiannis Demiris. 2017. "Quality and Diversity Optimization: A Unifying Modular Framework". IEEE Transactions on Evolutionary Computation 22 (2). IEEE:245–59.