Intelligent Agents Module - MSc AI Portfolio

Module Overview

This module explores the fundamental concepts and practical applications of intelligent agents in artificial intelligence. Through a combination of theoretical understanding and hands-on practice, we delve into various aspects of intelligent agents, including:

Agent architectures and design principles
Multi-agent systems and coordination
Learning and adaptation in agents
Reasoning and planning mechanisms
Real-world applications and case studies

Learning Outcomes

Understand the fundamental concepts of intelligent agents and multi-agent systems
Apply agent-based modeling techniques to real-world problems
Analyze and evaluate different agent architectures and coordination strategies
Develop effective solutions using agent-based approaches

Projects

Individual Project Advanced Agent Implementation Unit 11

An in-depth individual project focusing on the implementation and analysis of intelligent agents, demonstrating advanced concepts in agent design and coordination.

Team Project Multi-Agent System Design Unit 6

Collaborative team project focusing on designing and implementing a multi-agent system to solve complex coordination problems.

Discussions

Collaborative Discussion 1 Agent-Based Systems: Rise and Organisational Benefits Units 1-3

Exploration of agent-based systems and their benefits for organisations, examining how autonomous agents model complex problems, provide simulation tools for decision support, and solve decentralised control challenges.

Collaborative Discussion 2 Agent Communication Languages: Evolution and Adaptation Units 5-7

Analysis of agent communication languages (ACLs) like KQML, examining their advantages in enabling interoperability through speech acts, limitations from ontology dependence, and evolution toward learned, adaptive communication strategies.

Collaborative Discussion 3 Ethics of Deep Learning and Generative AI Units 9-11

Critical examination of ethical implications of deep learning and generative AI, including bias amplification, copyright issues, misinformation risks, and the need for governance frameworks like the EU AI Act and NIST standards.

e-Portfolio Activities

e-Portfolio Component Agent Dialogue with KQML and KIF Unit 6

Implementation of agent-to-agent communication using KQML and KIF protocols, demonstrating procurement agent (Alice) querying warehouse agent (Bob) about television inventory and specifications.

e-Portfolio Component Constituency-Based Parse Trees Unit 8

Analysis of natural language sentence structure using constituency-based parsing, demonstrating syntactic analysis and structural ambiguity resolution in intelligent agent language processing.

e-Portfolio Component Deep Learning and Societal Impact Unit 10

Critical analysis of Large Language Model agents, examining their technical capabilities and ethical, privacy, and socio-technical implications for society.

Final Evaluation & Reflective Analysis

Evaluation Intelligent Agents Module Evaluation Unit 12