0.25
0.5
0.75
1.25
1.5
1.75
2
Bootstrapping Skills
Published on Jul 28, 20151768 Views
The monolithic approach to policy representation in Markov Decision Processes (MDPs) looks for a single policy that can be represented as a function from states to actions. For the monolithic approac
Related categories
Chapter list
Bootstrapping Skills00:00
Outline00:15
Monolithic Policy00:42
Example: Monolithic Policy01:20
Skills01:51
Example: Skills02:29
Learning Skills03:22
Learning Skills via Bootstrapping (LSB)04:10
Model Iteration05:31
Experiment: Puddle World07:51
Experiments: Puddle World10:09
Experiments: Pinball - 111:04
Experiments: Pinball - 212:03
Conclusion12:31
Main Theorem14:30