📈 Stock Allocation using Reinforcement Learning (RL)

This project implements a multi-agent reinforcement learning (MARL) approach for stock portfolio allocation, enhanced with market regime awareness. The goal is to develop intelligent agents that learn to allocate capital efficiently across multiple stocks, dynamically adapting to different market regimes (e.g., bull, bear, sideways) for robust performance.

🚀 Project Highlights

• 🧠 Multi-Agent RL: Each agent learns to allocate funds to a subset or all of the assets, allowing for specialization and collaboration.
• 📊 Market Regime Awareness: Market regime detection models guide the agents to switch or adjust strategies based on current market conditions.
• 🔁 Continuous Learning: Agents learn from historical price data using advanced RL algorithms.

#Dataset Link (Google Drive) : https://drive.google.com/drive/folders/1O4pr29OoPD7zBEpe-N-8VSoGk8JtkzSL?usp=drive_link

Master Doc Link : https://docs.google.com/document/d/1OpM-dEcGLdNRB2YX-Cf1-FuFGoSICpkhWI5o7Eo4hwY/edit?tab=t.0