Independent project exploring how neural networks can evolve and adapt in fast-paced environment. Developed in C# (WinForms) from scratch as an experiment in evolutionary computation and agent-based modeling.
This project simulates a 2D environment populated by autonomous agents. Each organism is represented as a colored cell and is controlled by an artificial neural network that serves as its “brain.” Organisms interact with their surroundings — moving, photosynthesizing, attacking others, reproduce, consuming organic matter, etc. Through mutations and natural selection, more adaptive neural architectures gradually evolve over generations.
The aim is to observe emergent behavior, survival strategies, and how complex intelligence can arise from simple rules and mutation-driven evolution.
- Environment: Discrete 2D grid cellular field
- Agents: Each organism has parameters like energy, age, color, and neural network structure
- Actions: Move, photosynthesize, bite, consume organic matter, reproduce
- Neural Network: Defines decision-making; number of layers, neurons, and weights can mutate
- Evolution: Agents with higher fitness reproduce more often, passing slightly mutated networks to offspring
- Custom-built neural network system (no external ML libraries)
- Adjustable simulation parameters (mutation rate, resources amount, etc.)
- Different rendering modes(organics, species, etc.)
- Ability to view the structure of neural network
- User-friendly UI for pausing, observing, and tweaking parameters on the fly
| Simulation Overview | Neural Activity |
|---|---|
 |
 |
- Language: C#
- Framework: WinForms
- Paradigm: Object-Oriented Programming
- Concepts: Neural Networks, Genetic Algorithms, Artificial Life
- Presented at regional student project competition — 3rd Place Award
- Demonstrated at local science and technology events
- Developed independently as part of personal research on artificial evolution
- Implement speciation and cross-breeding
- Add visualization of neural network
- Introduce environmental variability (e.g., resource zones, temperature)
- Export and analyze long-term population data for research
To launch the project locally:
-
Install .NET Framework and ensure C# is supported on your system.
-
Clone the repository:
-
Open the solution file (.sln) in Visual Studio (or any IDE that supports C# and WinForms).
-
Build and run the project — the simulation window will appear automatically.