About

Welcome to Digital organisms ecology system. This is an optimized 2D ecosystem with the ability to run up to 10 000 agents (organisms) and 100 000 molecules (matter and organism's food) in parallel. It uses linear assembler-like language (LGP)as a DNA, which runs under self-made VM. The purpose is to produce complex behaviour of organisms based on generated code. Such code is used as an organism's "brain" and manages all his life aspects. Also, this system uses evolution to change (mutate) and select (natural selection) best agents from the population in the same way as nature does. This is how the mentioned complex behaviour should appear. Through generations, every new organism should, in theory, be more adaptive to the virtual environment and more complex (but not always). Instead of neural networks, organisms use special byte code (mentioned "brain"), which run in an infinite loop. Mutations affect this code to produce small variations in population. Best organisms will have more chances to produce children. This is, by the way, major criteria for such a system. Main ideas obtained from Synthesis theory and Self-replicating machines. The 2D world consists of organisms and atoms - code instructions, which are food for constructing organisms bodies. Enjoy our research!

Benefits

The project gives you an opportunity to peek into the lives of digital organisms and observe them in real-time mode. Here is a quick demo to get the general idea of this digital yet living environment. The project is quite complex in regard to its consequences and unpredictability of the final results as well as its evolutionary essence. The patterns identified in-here could be later used to advance on the understanding or improving machine learning, artificial intelligence and many more.

Now, we are looking for an enthusiast who would contribute to the project and develop it further. If you've got interested, surely contact us!