sugarscape

KO | EN

gitlite — search

#pdf #proxy #api #web #docker #large-language-models #tailwindcss #data-science #agent #gpt

(★ 9)

A Java implementation of the Sugarscape agent-based simulation model, featuring complex demographics, decentralized trade, financial credit systems, and epidemiology.

파일 탐색기

ZIP 다운로드

- Config.java
- DefaultConfig.java
- Display.java
- IDisplay.java
- Menu.java
- IWorldInitializer.java
- WorldInitializer.java
- Agent.java
- BimodalGaussianDistribution.java
- Contract.java
- Disease.java
- IAgeCapabilities.java
- IAgent.java
- IContract.java
- IDisease.java
- IInfectable.java
- ILocatable.java
- IMapGenerator.java
- IMetabolic.java
- IPatch.java
- IReproducible.java
- IResource.java
- ITradable.java
- IWorld.java
- ModelFactory.java
- Patch.java
- ResourceType.java
- Spice.java
- Sugar.java
- World.java
- AgentUtilities.java
- GrowBack.java
- IGrowBack.java
- IImmunity.java
- IInfection.java
- ILoan.java
- IMigration.java
- Immunity.java
- Infection.java
- IReproduce.java
- ITrade.java
- Loan.java
- Migration.java
- Reproduce.java
- Trade.java
- AgeStep.java
- ImmunityStep.java
- InfectionStep.java
- ISimulationEngine.java
- ISimulationStep.java
- LoanStep.java
- MigrationStep.java
- RegrowthStep.java
- ReproductionStep.java
- SimulationEngine.java
- TradeStep.java
Main.java
README.md

# CDN으로 사용하기

jsDelivr

jsDelivr는 공개 GitHub 리포지토리를 별도 설정 없이 CDN으로 즉시 서빙합니다. 버전과 파일을 고르면 웹페이지에 바로 붙일 수 있는 링크와 예시 코드가 만들어집니다.

버전

파일

링크

예시

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

Welcome to the **Sugarscape Simulation** project. This repository contains the Java implementation of the Sugarscape model, culminating in a fully integrated application.

Sugarscape is an agent-based social and economic simulation model. It creates a silicon-based society to study emergent behaviors such as migration patterns, population dynamics, wealth inequality, trade, financial credit systems, epidemiology, and the conditions for societal survival.

---

## 📖 Theoretical Background

The model is based on the pioneering work of **Joshua M. Epstein & Robert Axtell** in their book [*Growing Artificial Societies*](https://www.google.com/search?q=Growing+Artificial+Societies+Epstein+Axtell). The roots of agent-based modeling can be traced back to:
*   **Thomas Schelling (1969)**: Segregation models that showed how local choices lead to global segregation.
*   **John Conway (1990s)**: The *Game of Life*, which popularized cellular automata.
*   **Epstein & Axtell**: Developed the Sugarscape model as a laboratory for social sciences, modeling how agents interact on a 2-D landscape, harvest resources, reproduce, trade, contract diseases, and borrow wealth.

---

## 🌍 Core Environment & Agent Design

### The Grid (Environment)
*   **Size**: A flat $51 \times 51$ grid divided into squares called **patches**.
*   **Resources**: Each patch contains a current amount of `sugar` (and later `spice`), alongside a `maximumCapacity`.
*   **Initial Setup**: Initial resource levels are assigned using a probability distribution.
*   **UI Representation**: 
    *   **Phases 1-3**: Patches visualize sugar using a color gradient from White ($0$) to Dark Orange (maximum).
    *   **Phases 4-8**: Patches are split vertically. The left half displays spice (White to Purple gradient), and the right half displays sugar (White to Orange gradient).
    *   **Agents**: Visualized as a dot. In early phases, they are solid red dots. In later phases, color indicates gender (Red/Blue) and fertility status (Gray if past reproductive age).

### The Agents
Agents occupy patches and have both fixed genetic traits and dynamic state variables:
1.  **Wealth**: The agent's current `sugar` and `spice` reserves.
2.  **Initial Endowment**: The amount of resources given to an agent at birth.
3.  **Metabolism**: Genetic traits dictating how many resources the agent consumes per tick to survive.
4.  **Vision**: Genetic trait determining the distance an agent can see in cardinal directions to migrate.
5.  **Demographics**: Variables like `age`, `maxAge`, `gender`, and `fertileLimits` that govern life cycles and reproduction.

---

## ⚡ Simulation Phases

This application integrates all 8 developmental phases of Sugarscape into a single project. Upon launching, a **GUI Menu** allows you to select which phase to execute. The rules activate incrementally based on your selection:

### 📅 Phase 1: Migration & Resource Growback
Introduces basic migration and resource growback.
*   **Growback**: At each tick, resources replenish up to their maximum capacity.
*   **Migration**: Agents evaluate empty patches within their vision, move to the one with the highest resources (closest in case of a tie), and harvest it.

### 🌊 Phase 2: Migration & Emergence (Herd Behavior)
Modifies parameters to trigger wave-like group migrations.
*   Agents are clustered initially in a dense $40 \times 40$ area in the bottom-left corner.
*   Vision is increased and growback interval is slowed down, resulting in diagonal wave patterns migrating across the Sugarscape.

### 🧬 Phase 3: Demographics & Reproduction
Introduces lifespans, genders, and sexual reproduction.
*   Agents age and eventually die of old age.
*   **Reproduction**: Fertile agents of opposite genders on adjacent patches can mate if they have accumulated enough wealth. They pass down a portion of their wealth to their offspring, who inherits a mix of their genetic traits (vision and metabolism).

### 🌶️ Phase 4: Two Resources (Sugar & Spice)
Adds a second critical resource, **spice**, creating a two-resource economy.
*   Agents must balance harvesting both sugar and spice, evaluated using the Cobb-Douglas welfare function.
*   If an agent runs out of either resource, they perish.

### 🤝 Phase 5: Trade & Decentralized Markets
Introduces bilateral trade between agents to prevent resource starvation.
*   **Marginal Rate of Substitution (MRS)**: Represents the agent's valuation of sugar relative to spice.
*   Agents interact with their neighbors to negotiate trades. Prices are determined by the geometric mean of their MRS values. 
*   Trading repeats until all mutually beneficial exchanges have cleared.

### 💳 Phase 6: Financial Credit & Loans
Allows lending and borrowing to facilitate reproduction for resource-poor agents.
*   Fertile agents without enough resources to reproduce can take out loans with a fixed interest rate.
*   Lenders are either wealthy infertile agents or fertile agents with a massive surplus. 
*   Unpaid debts are rolled over with compounding interest, and deaths cancel the contract.

### 🦠 Phase 7: Epidemiology & Immunology
Introduces disease propagation and immune defense mechanisms.
*   Diseases and immune systems are represented as binary strings. An agent is immune if the disease string is a substring of their immune system.
*   Infections increase metabolic costs, simulating sickness.
*   Agents gradually adapt to diseases by mutating their immune strings. Episodic outbreaks periodically hit the population.

### 💀 Phase 8: The Edge of Life (System Integration)
Integrates all rules and parameters into a single ecological simulation.
*   **Environmental Handicap**: To reflect the harsh reality of living with two resources and all systemic rules, the agents' maximum lifespan is strictly reduced.
*   This phase demonstrates how the combination of trade, loans, and immunity allows artificial societies to survive against steep environmental and biological odds.

---

## 🛠️ Setup & Running Instructions

### Prerequisites
*   Java Development Kit (JDK) 11 or higher recommended.

### Compiling and Running
To launch the integrated application, compile the source files and run the `Main` class. The application will open a graphical menu allowing you to select the phase you wish to simulate.

```bash
# Compile all Java files
javac $(find . -name "*.java")

# Run the simulation launcher
java Main
```

*(Note: On Windows PowerShell, use `Get-ChildItem -Recurse -Filter *.java | ForEach-Object { javac $_.FullName }` or simply compile from an IDE like IntelliJ / Eclipse.)*

sugarscape

파일 탐색기

# CDN으로 사용하기

Was this content helpful?

Recent Feedback

Download README