COMP6460 Reinforcement Learning and Planning Under Uncertainty
COMP6460 is only available under certain award programs.
| Offered By | Department of Computer Science |
|---|---|
| Academic Career | Graduate Coursework |
| Course Subject | Computer Science |
| Offered in | COMP6460 will not be offered in 2009 |
| Unit Value | 6 units |
| Course Description | This course provides an introduction to reinforcement learning (RL) and planning under uncertainty, thereby providing concepts for understanding and developing intelligent systems. For instance, the world-class Backgammon program, TD-Gammon, is based on RL techniques. Topics covered will be the classical MDP model, temporal difference learning, dynamic programming, structured models, apporximation algorithms, integrating planning and learning, and the theory of universal rational agents based on sequential decision theory and algorithmic information theory. |
| Indicative Assessment | Two written assingments (15% each); Written Examination (70%) |
| Workload | Thirteen three hour lectures |
| Course Classification(s) | AdvancedAdvanced courses are designed for students having reached 'first degree' level of assumed knowledge, which provide a deep understanding of contemporary issues; or 'second degree' and higher levels of knowledge; or for transition to research training programs. |
| Areas of Interest | Computer Science |
| Requisite Statement | Enrolment in the MICT or approval by the program convenor |
| Recommended Courses |
COMP6320 Artificial Intelligence and /or COMP6467 Introduction to Statistical machine Learning |
| Consent Required | Consent is required prior to enrolling in this course. |
The information published on the Study at ANU 2009 website applies to the 2009 academic year only. All information provided on this website replaces the information contained in the Study at ANU 2008 website.
