ECE3811 · Reinforcement learning for engineers

Overview

This unit introduces you to the foundational principles and practice of reinforcement learning and its application to sequential decision-making problems in complex dynamic systems. Topics include reinforcement learning fundamentals (Markov decision process, Bellman equation, value and policy iteration), Q-learning, actor-critic methods, policy gradient, and imitation learning. You will also examine practical approaches for implementing these methods, including deep learning-based reinforcement learning, exploration-exploitation strategies, and reward function design. Throughout the course, you will learn how these methods are applied to solve planning, control, and optimisation problems in simulated engineering environments, and develop the necessary skills to enable modern intelligent decision-making systems.

Offerings

Assessment

The Handbook lists an examination for this unit.

Continuous assessment: 50% Final assessment: 50% Assessment in this unit includes competency hurdle assessment task/s. The consequence of not achieving a competency hurdle is a fail grade (NH) and a maximum mark of 45 for the unit.

Assessment details may change. Please refer to the assessment information in Moodle closer to the start of the teaching period.

Requisites

Learning outcomes

1. Apply appropriate reinforcement learning algorithms with representative state and action spaces to sequential decision-making problems.
2. Design model-based and model-free reinforcement learning solutions that meet specified performance requirements.
3. Demonstrate independent learning by using modern simulation platforms to implement decision-making algorithms.

Workload

The minimum total expected workload to achieve the learning outcomes for this unit is 144 hours per semester, typically comprising a mixture of 3-6 hours of scheduled learning activities and 6-9 hours of independent study per week. Scheduled activities may include a combination of teacher-directed learning, peer-directed learning and online engagement. Independent study may include associated readings, assessment and preparation for scheduled activities.

Student experience, not official rules. Nothing here changes what the Handbook says.