Please use this identifier to cite or link to this item: https://scholarbank.nus.edu.sg/handle/10635/212722
Title: DIFFERENTIABLE ROBOTICS: COMPOSITIONAL DEEP LEARNING WITH DIFFERENTIABLE ALGORITHM NETWORKS
Authors: PETER KARKUS
ORCID iD:   orcid.org/0000-0002-1474-9771
Keywords: deep learning, planning, model-based, model-free, end-to-end, modular, neural network
Issue Date: 1-Aug-2021
Citation: PETER KARKUS (2021-08-01). DIFFERENTIABLE ROBOTICS: COMPOSITIONAL DEEP LEARNING WITH DIFFERENTIABLE ALGORITHM NETWORKS. ScholarBank@NUS Repository.
Abstract: Towards human-level robot intelligence a central question is the architecture of robot learning: how is the system represented, and how is it trained? This thesis introduces the Differentiable Algorithm Network (DAN), a compositional architecture for designing robot learning systems. The DAN is composed of neural network modules, each encoding an algorithm and associated models; and it is trained end-to-end from data. The key idea is to make model-based algorithms differentiable and encode them in a generalized neural network, thus combining the model-based modular system design with data-driven end-to-end learning. Algorithms act as structural assumptions to moderate data requirements; end-to-end learning allows modules to adapt and compensate for imperfections. We introduce DANs for a range of domains: particle filter networks for visual localization; differentiable SLAM networks for visual SLAM; QMDP networks for partially observable planning. We also combine DAN modules for visual navigation in simulation and with a real-world quadruped robot.
URI: https://scholarbank.nus.edu.sg/handle/10635/212722
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
KarkusP.pdf7.66 MBAdobe PDF

OPEN

NoneView/Download

Page view(s)

54
checked on Sep 22, 2022

Download(s)

18
checked on Sep 22, 2022

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.