Abstract
This paper develops a new method for coevolution, named Fitness-Diversity Driven Coevolution (FDDC). This approach builds on existing methods by a combination of a (predator-prey) Coevolutionary Genetic Algorithm (CGA) and novelty search. The innovation lies in replacing the absolute novelty measure with a relative one, called Fitness-Diversity. FDDC overcomes problems common in both CGAs (premature convergence and unbalanced coevolution) and in novelty search (construction of an archive). As a proof of principle, Spring Loaded Inverted Pendulums (SLIPs) are coevolved with 2Dterrains the SLIPs must learn to traverse.
Original language | English |
---|---|
Title of host publication | IEEE Congress on Evolutionary Computation |
Publisher | IEEE |
Pages | 201-208 |
Number of pages | 8 |
ISBN (Print) | 9781509046010 |
DOIs | |
Publication status | Published - 2017 |
Event | IEEE Congress on Evolutionary Computation (CEC) - San Sebastián, Spain Duration: 5 Jun 2017 → 8 Jun 2017 https://www.esteco.com/corporate/ieee-congress-evolutionary-computation-cec-2017#:~:text=CEC%202017%20is%20a%20congress,for%20the%20benefit%20of%20humanity. |
Publication series
Series | IEEE Congress on Evolutionary Computation |
---|
Conference
Conference | IEEE Congress on Evolutionary Computation (CEC) |
---|---|
Abbreviated title | CEC 2017 |
Country/Territory | Spain |
City | San Sebastián |
Period | 5/06/17 → 8/06/17 |
Internet address |
Keywords
- LOADED INVERTED PENDULUM
- OPTIMIZATION