0.25
0.5
0.75
1.25
1.5
1.75
2
Robots learning from human teachers
Published on Jul 28, 20153790 Views
In this talk I present recent work from the Socially Intelligent Machines Lab at Georgia Tech. The vision of our research is to enable robots to function in real human environments; such as, service r
Related categories
Chapter list
Robots Learning from Human Teachers 00:00
Social Robot00:01
Robot Successes00:16
Our Goal - 100:46
Our Goal - 201:06
Our Goal - 301:45
Interactive Machine Learning - 102:15
Interactive Machine Learning - 203:43
Interactive Machine Learning - 304:12
Interactive Machine Learning - 404:26
Interactive Reinforcement Learning - 104:37
Interactive Reinforcement Learning - 204:47
Reward Shaping04:57
Sophie’s Kitchen Results05:25
Policy Shaping06:02
Human’s Feedback Policy06:59
Agent’s Policy08:04
Policy Shaping: ADVISE08:19
Experiments - 108:55
Experiments - 209:19
Experiments - 309:29
Experiments - 409:45
Interactive Machine Learning - 510:23
Interactive Machine Learning - 610:52
Interactive Machine Learning - 711:49
Demo - 111:52
Trajectory Demonstrations12:17
Demo - 212:55
Keyframe Demonstrations13:47
Demo - 314:06
Trajectory vs. Keyframe Demonstrations - 114:57
Trajectory vs. Keyframe Demonstrations - 215:13
Keyframe-based Learning from Demonstration - 115:30
Keyframe-based Learning from Demonstration - 215:41
Keyframe-based Learning from Demonstration - 316:07
Embodied Queries for Robot Learners16:46
Generating Questions - 117:04
LABEL QUERIES18:03
Generating Questions - 218:32
FEATURE QUERIES19:14
Interactive Machine Learning - 820:16
Demo - 420:56
Learning Goals from Demonstration22:43
Visual Goals23:31
Haptic Goals23:50
Pose Constraint Goals24:07
Learning Goals from Demonstration - 124:24
Learning Goals from Demonstration - 224:52
Haptic Goals: Action-Object Affordances24:56
Demo - 525:07
Learning Haptic Affordances25:12
Experiment25:38
Action Demonstration25:51
Human Guided Exploration26:43
Affordance Modeling27:07
Results: Online Affordance Testing - 127:41
Results: Online Affordance Testing - 227:45
Results: Online Affordance Testing - 327:59
Results: Online Affordance Testing - 428:29
Learning Goals from Demonstration28:59
Best of Both: Keyframe and Trajectory - 129:52
Best of Both: Keyframe and Trajectory - 230:17
Best of Both: Keyframe and Trajectory - 330:28
Keyframe Demo for Geometric Constraint30:53
Trajectory Demo for Dynamic Constraint31:12
Resulting planned motion with constraints31:33
Interactive Machine Learning - 832:15
Robots Learning from Human Teachers32:17
Socially Intelligent Machines Lab32:19
Questions?32:23