Traffic Routing with Optimal Path in Graph-Based Networks - Deep Reinforcement Learning

Introduction

This repository contains a Deep Reinforcement Learning (DRL) approach to optimize traffic routing in graph-based networks. Unlike conventional algorithms, our model considers a broader range of network parameters to minimize delay and congestion efficiently. The model is formulated with a custom environment, reward function, and agent capable of learning an optimal policy for routing.

Features

• Utilizes Deep Reinforcement Learning to optimize traffic routing in graph-based networks.
• Overcomes limitations of conventional algorithms by considering a wider range of network parameters.
• Customizable environment and reward function to suit specific network scenarios.
• Agent is designed to learn an optimal policy for routing over time.