University of Galway

Course Module Information

Course Modules

Semester 2 | Credits: 5

The topic of Agents and Multi-Agent Systems, examines environment that involve autonomous decision making software actors to interact with their surroundings with the aim of achieving some individual or overall goal. A typical agent environment could be a trading environment where an agent attempts to optimise energy usage, or the profitability of a transaction. More recently, significant global attention has focussed on the vision of autonomous vehicles, which also follows the core principle of an agent attempting to achieve a set of defined goals. This module begins by examining the underpinnings of what is an Agent, and how we can better understand the principles of an agent and its autonomy. Multi-Agent Systems are then explored, as a means of understanding how many agents can interact with each other in a complex environment. Agents are commonly modelled using Game Theory, and in this module a range of Game Theoretic Models will be studied. The module will examine Adaptive Learning Agents through the use of Reinforcement Learning algorithms an area of Machine Learning, which focuses on training learners to choose actions which yield the maximum reward in the absence of prior knowledge. The module takes a hands-on, practical approach to reinforcement learning theory, beginning with Markov Decision Processes, detailing practical learning examples in discrete environments and how to formulate a reinforcement learning task. It then extends this to continuous problem spaces, detailing Deep Reinforcement Learning with a practical implementation of a Deep Q Network using Keras.
(Language of instruction: English)

Learning Outcomes
  1. Explain and discuss the principles underlying Agents.
  2. Explain the role of game theory and games in agent design.
  3. Apply the principle of agents to a range of simulation problems.
  4. Understand the theory unpinning reinforcement learning.
  5. Apply reinforcement learning to a real-world problem.
  6. Apply advanced deep reinforcement learning approaches to a real-world problem.
Assessments
  • Written Assessment (70%)
  • Continuous Assessment (30%)
Teachers
The above information outlines module CT5134: "Agents, Multi-Agent Systems and Reinforcement Learning " and is valid from 2019 onwards.
Note: Module offerings and details may be subject to change.