0.25
0.5
0.75
1.25
1.5
1.75
2
Finite horizon exploration for path integral control problems
Published on Feb 25, 20074859 Views
We have recently developed a path integral method for solving a class of non-linear stochastic control problems in the continuous domain [1, 2]. Path integral (PI) control can be applied for timede
Related categories
Chapter list
Finite horizon exploration for path integral control<br> problems00:01
Stochastic optimal control theory00:51
Exploration in RL03:03
Exploration in receding horizon problems04:46
Finite horizon exploration04:59
Outline06:07
Stochastic optimal control06:32
The optimal cost-to-go08:19
The diffusion process09:52
The path integral formulation10:45
An example: double slit11:40
MC sampling on double slit12:02
Planning: RL vs PI control12:24
Planning: RL vs PI control0113:14
Some observations13:48
Fixed horizon exploration14:21
Fixed horizon exploration0116:49
Fixed horizon exploration0217:07
Fixed horizon exploration0317:49
Fixed horizon exploration0418:10
Fixed horizon exploration0519:00
Receding horizon exploration20:23
Receding horizon exploration0122:11
Receding horizon exploration0223:49
Summary24:06
Further reading25:25