Learning perception and control from rich interactions
Abstract/Contents
- Abstract
- Building robotic systems that can perform a wide range of tasks in the real world would require generalizable perception and control. Despite recent advances in deep learning, existing paradigms often rely on extensive human engineering and suffer from limited generalization capability. To handle the desired diversity and complexity in challenging tasks, we would need to further scale up learning by rethinking the methodology for collecting and utilizing data for robots. In this dissertation, we discuss methods that enable robots to enhance perception and control by learning from rich interactions with the environment. We develop structured models and learning algorithms for robots to effectively acquire sensorimotor skills and solve sequential tasks. To scale up learning, we design mechanisms to provide data from various sources. We further propose frameworks that adaptively generate tasks from parameterized task spaces to facilitate curriculum learning and skill discovery.
Description
Type of resource | text |
---|---|
Form | electronic resource; remote; computer; online resource |
Extent | 1 online resource. |
Place | California |
Place | [Stanford, California] |
Publisher | [Stanford University] |
Copyright date | 2021; ©2021 |
Publication date | 2021; 2021 |
Issuance | monographic |
Language | English |
Creators/Contributors
Author | Fang, Kuan |
---|---|
Degree supervisor | Li, Fei Fei, 1976- |
Degree supervisor | Savarese, Silvio |
Thesis advisor | Li, Fei Fei, 1976- |
Thesis advisor | Savarese, Silvio |
Thesis advisor | Guibas, Leonidas J |
Thesis advisor | Sadigh, Dorsa |
Degree committee member | Guibas, Leonidas J |
Degree committee member | Sadigh, Dorsa |
Associated with | Stanford University, Department of Electrical Engineering |
Subjects
Genre | Theses |
---|---|
Genre | Text |
Bibliographic information
Statement of responsibility | Kuan Fang. |
---|---|
Note | Submitted to the Department of Electrical Engineering. |
Thesis | Thesis Ph.D. Stanford University 2021. |
Location | https://purl.stanford.edu/gn280rg9424 |
Access conditions
- Copyright
- © 2021 by Kuan Fang
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...