Dominated Novelty Search: Rethinking Local Competition in Quality-Diversity

This repository contains the official implementation of "Dominated Novelty Search: Rethinking Local Competition in Quality-Diversity".

Overview

Dominated Novelty Search (DNS) is a novel Quality-Diversity algorithm that implements local competition through dynamic fitness transformations, eliminating the need for predefined bounds or parameters. Our method:

• Outperforms existing approaches across standard Quality-Diversity benchmarks
• Maintains high performance in high-dimensional and unsupervised spaces
• Serves as a drop-in replacement for grid mechanisms in MAP-Elites

Installation

Requirements

• Python 3.8+
• CUDA-compatible GPU (recommended)

Quick Start

# Clone the repository
git clone https://github.com/adaptive-intelligent-robotics/Dominated-Novelty-Search
cd Dominated-Novelty-Search

# Create and activate virtual environment (optional but recommended)
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt