XPERIENCE - Robots Bootstrapped through Learning from Experience thumbnail
slide-image
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

XPERIENCE - Robots Bootstrapped through Learning from Experience

Published on Mar 14, 20122983 Views

Current research in enactive, embodied cognition is built on two central ideas: 1) Physical interaction with and exploration of the world allows an agent to acquire and extend intrinsically grounded,

Related categories

Chapter list

XPERIENCE - Robots Bootstrapped through Learning from Experience00:00
Xperience: Problem and Approach00:21
Main Novelty of Xperience02:32
Structural Bootstrapping02:51
Examples for Structural Bootstrapping04:24
Major Scientific Questions05:30
OACs as representations in Xperience06:50
The XPERIENCE Cognitive Architecture08:27
OACs on all levels09:51
Development and Structural Bootstrapping10:53
Learning hierarchical and probabilistic sensory-motor spaces11:49
Machine Learning techniques for exploration-based ...12:41
Finding Structure in Objects x Features x Actions12:59
Generalizing Objects by Analyzing Language ("GOAL") - 113:33
Generalizing Objects by Analyzing Language (“GOAL”) - 214:28
Pushing reflex for learning object representations15:27
Switching Motor Primitives in Collaborative Tasks15:57
Tightly-coupled physical human-robot interaction16:32
Language and planning domain16:59
Scenario: "Human Living Space"17:32
Robot Platforms in Xperience18:10
Thank you for your attention!18:27