Multi-Agent Systems & Coordination in Python Training Course

Introduction to Multi-Agent Systems

• Overview of agents, environments, and interaction models.
• Exploring cooperation, competition, and autonomy within agentic systems.
• Applications across logistics, robotics, and decision-making domains.

Core Concepts of Agent Architecture

• Distinguishing between reactive and deliberative agents.
• Examining communication protocols and coordination models.
• Knowledge representation and managing shared state.

Implementing Agents in Python

• Constructing agents using the Mesa framework.
• Modelling environments and agent interactions.
• Simulating agent behavior and generating visualisations.

Coordination and Communication

• Architectures for message passing and shared memory.
• Processes for negotiation, consensus building, and task allocation.
• Coordination algorithms, including contract net, market-based, and swarm models.

Learning and Adaptation in Multi-Agent Systems

• Applying reinforcement learning to multiple agents.
• Analyzing cooperative versus competitive learning dynamics.
• Utilising PettingZoo and Stable-Baselines3 for Multi-Agent Reinforcement Learning (MARL).

Distributed Computing and Scaling

• Leveraging Ray for distributed multi-agent simulations.
• Managing concurrency and synchronization effectively.
• Parallelizing computation and handling shared resources.

Human–Agent Collaboration

• Designing interfaces for human-in-the-loop coordination.
• Creating hybrid workflows with AI-assisted decision support.
• Considering ethical and operational implications.

Capstone Project

• Design and implement a comprehensive multi-agent system in Python.
• Demonstrate coordination and learning capabilities among agents.
• Present simulation results and key performance insights.

Summary and Next Steps