Search

This poster describes our approach to procedurally generating video game creatures through the evolution of digital genomes. The creature genomes are modeled in the Unity game engine using floating point variables to represent genes. The genetic values are then used to calculate game specific traits that can be tied to fitness functions during game design.

This poster describes our approach to procedurally generating video game creatures through the evolution of digital genomes. The creature genomes are modeled in the Unity game engine using floating point variables to represent genes. The genetic values are then used to calculate game specific traits that can be tied to fitness functions during game design.

Darwin’s Demons is a video game that incorporates evolutionary biology. It features a population of enemies encoded by diploid digital genomes of quantitative trait loci. Each round of gameplay is modeled as a generation in which a new enemy population is produced using tournament selection with replacement. Each parent provides a chromosome produced with free recombination. Mutations are then drawn from a Gaussian distribution. Enemies gain fitness by shooting close to the player and spending time near the bottom of the screen. These conditions produce rapid rates of adaptation to player strategies. Ultimately, Darwin’s Demons allows players to observe (and sometimes exploit) evolution through the course of gameplay.

Developmental robotics (also known as epigenetic robotics) is mainly concerned with modeling the postnatal development of cognitive behaviors in living systems, such as language, emotion, curiosity, anticipation, and social skills. While current work in this field has shown significant successes, we believe integrating research on developmental (including epigenetic and morphogenetic) robotics and evolutionary robotics is the natural next step. This workshop aims at bringing together evolutionary robotics and developmental robotics to form a new research area "evolutionary developmental robotics" (evo-devo-robo). The present paper contains the abstracts of the talks given by each of the seven invited speakers. These abstracts cover research in both fields and give an overview of the potential interactions between developmental and evolutionary robotics.

For a given data set, its set of attributes defines its data space representation. The quality of a data space representation is one of the most important factors influencing the performance of a data mining algorithm. The attributes defining the data space can be inadequate, making it difficult to discover high-quality knowledge. In order to solve this problem, this paper proposes a Genetic Programming algorithm developed for attribute construction. This algorithm constructs new attributes out of the original attributes of the data set, performing an important preprocessing step for the subsequent application of a data mining algorithm.