Abstract | ||
---|---|---|
In Reinforcement Learning (RL), it is common to use optimistic initialization of value functions to encourage exploration. However, such an approach generally depends on the domain, viz., the scale of the rewards must be known, and the feature representation must have a constant norm. We present a simple approach that performs optimistic initialization with less dependence on the domain. |
Year | Venue | Field |
---|---|---|
2014 | AAAI Workshop: Learning for General Competency in Video Games | Mathematical optimization,Computer science,Artificial intelligence,Initialization,Machine learning,Reinforcement learning |
DocType | Volume | Citations |
Journal | abs/1410.4604 | 5 |
PageRank | References | Authors |
0.51 | 1 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Marlos C. Machado | 1 | 135 | 14.48 |
Sriram Srinivasan | 2 | 379 | 27.92 |
Michael H. Bowling | 3 | 2460 | 205.07 |