Self-evolution in a constructive binary string system
Artificial Life
Fifty years of research on self-replication: an overview
Artificial Life - Special issue on self-replication
Evolving control metabolism for a robot
Artificial Life
Artificial chemistries—a review
Artificial Life
Genetic Programming and Autoconstructive Evolution with the Push Programming Language
Genetic Programming and Evolvable Machines
The Push3 execution stack and the evolution of control
GECCO '05 Proceedings of the 7th annual conference on Genetic and evolutionary computation
Efficient operation in sensor and actor networks inspired by cellular signaling cascades
Proceedings of the 1st international conference on Autonomic computing and communication systems
GECCO '96 Proceedings of the 1st annual conference on Genetic and evolutionary computation
Bio-Inspired Approaches for Autonomic Pervasive Computing Systems
Bio-Inspired Computing and Communication
Hi-index | 0.00 |
We present PlasmidPL, a plasmid-inspired programming language designed for Genetic Programming (GP), and based on a chemical metaphor. The basic data structures in PlasmidPL are circular virtual molecules or rings which may contain code and data. Rings may react with each other to perform computations on the rings themselves. A virtual chemical reactor stochastically chooses which reactions should occur and when. Code and data may be rewritten in the process, leading to a system that constantly modifies itself. In order to be closer to chemistry, PlasmidPL relies solely on the data and code stored in molecules. After describing the language, we show some hand-written sample programs that implement initial program generation, mutation and crossover within self-modifying chemical programs. These programs are then used to solve a typical symbolic regression problem, as a feasibility study. Finally, we discuss future directions into specific application scenarios that can benefit from such a chemical model.