COMP6008

COMP6008: Reinforcement Learning

Science and Engineering

Student reviews for COMP6008 at Curtin University. See ratings, workload, grades, and honest feedback from Curtin students.

View in Curtin Handbook

About this Unit

This unit will provide introduction to reinforcement learning and relevant practical background to solve challenging real-world problems. It covers both model-free and model-based approaches. Students will study tabular reinforcement learning methods such as multi-armed bandits, dynamic programming, temporal-difference learning including Q-learning and SARSA. Different policy search techniques, such as policy gradient methods, are discussed. Finally, the unit also covers advanced reinforcement learning topics related to deep learning and practical considerations.

Quick Tips

One-line things worth knowing. No full review required.

No tips yet. Be the first to share one.

Unit Details

Credits25

Contact Hours3 Weekly

Result TypeGrade/Mark

AreaSchool of Electrical Engineering, Computing and Mathematical Sciences (EECMS) (1153)

Field of Education020100 Computer Science (Narrow Grouping)

Handbook View Official Handbook Page →

Tuition Pattern

Workshop1 x 3 hours weekly

Prerequisites

Prerequisites group 1Select One

MC-AINTLv3Master of Artificial Intelligence
MC-MATHSCv3Master of Science (Mathematical Sciences)

Prerequisites group 2Select One

Select One

COMP5005v1Fundamentals of Programming

Select One

Student Reviews

No reviews yet. Be the first to write one!