Library Open Repository
Reducing the Time Complexity of Goal-Independent Reinforcement Learning
Ollington, R and Vamplew, P (2004) Reducing the Time Complexity of Goal-Independent Reinforcement Learning. In: AISAT2004: International Conference on Artificial Intelligence in Science and Technology, 21-25 November 2004, Hobart, Tasmania, Australia.
AISAT-CQL.pdf | Download (407kB)
Available under University of Tasmania Standard License.
Concurrent Q-Learning (CQL) is a goal independent reinforcement learning technique that learns the action values to all states simultaneously. These action values may then be used in a similar way to eligibility traces to allow many action values to be updated at each time step. CQL learns faster than conventional Q-learning techniques with the added benefit of being able to apply all experiences gained performing one task to any new task within the problem domain. Unfortunately the update time complexity of CQL is O(|S|2x|A|). This paper presents a technique for reducing the update complexity of CQL to O(|A|) with little impact on performance.
|Item Type:||Conference or Workshop Item (Paper)|
|Keywords:||goal-independent reinforcement learning, hierarchical reinforcement learning|
|Page Range:||pp. 132-137|
|Date Deposited:||26 Nov 2004|
|Last Modified:||18 Nov 2014 03:10|
|Item Statistics:||View statistics for this item|
Repository Staff Only (login required)
|Item Control Page|