A-MCTS: Adaptive Monte Carlo Tree Search for Temporal Path Discovery

An implementation of Monte Carlo Tree Search with reinforcement learning for finding optimal paths in temporal networks.

• 📄 Read the Paper
• 🌐 Live Demo

Overview

A-MCTS efficiently discovers temporal paths in complex networks with multiple constraints, applicable to transportation, communication systems, and social network analysis.

Quick Start

# Install dependencies
conda env create -f environment.yml  # or pip install -r requirements.txt
conda activate amcts

# Run locally
python app.py  # Then navigate to http://localhost:5000

Repository Contents

• src/: Core algorithm implementation
• data/: Sample datasets
• notebooks/: Examples and demonstrations
• app.py: Web application

Citation

P. Ding, G. Liu, Y. Wang, K. Zheng and X. Zhou, "A-MCTS: Adaptive Monte Carlo Tree Search for Temporal Path Discovery," in IEEE Transactions on Knowledge and Data Engineering, vol. 35, no. 3, pp. 2243-2257, 1 March 2023, doi: 10.1109/TKDE.2021.3120993. keywords: {Heuristic algorithms;Monte Carlo methods;Vehicle dynamics;Costs;Memory management;Estimation;Urban areas;Temporal path;monte carlo tree;path finding}