Deep Learning for Control: a non-Reinforcement Learning View

被引：0

作者：

Matei, Ion ^{[1
]}

Minhas, Raj ^{[1
]}

Zhenirovskyy, Maksym ^{[1
]}

de Kleer, Johan ^{[1
]}

Rai, Rahul ^{[2
]}

机构：

[1] Palo Alto Res Ctr Inc PARC, Palo Alto, CA 94304 USA

[2] Univ Buffalo SUNY, Buffalo, NY USA

来源：

2020 AMERICAN CONTROL CONFERENCE (ACC) | 2020年

关键词：

D O I：

10.23919/acc45564.2020.9147287

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Deep learning platforms have become hugely popular due to their successes in natural language processing and image processing. Our objective is to show how deep learning platforms can be used for control problems. We do not make judgments about their performance as compared to traditional control approaches. We show that the main challenge when using deep learning platforms for learning control policies for nonlinear systems is ensuring the stability of the learning algorithm that depends on the stability of the closed loop system during the learning process. We discuss two approaches for overcoming the potential instability of the optimization algorithm, and showcase them in the context of learning a stabilizing controller for an inverted pendulum.

引用

页码：2942 / 2948

页数：7

共 50 条

[31] Satellite Attitude Control with Deep Reinforcement Learning
Gao, Duozhi
Zhang, Haibo
Li, Chuanjiang
Gao, Xinzhou
2020 CHINESE AUTOMATION CONGRESS (CAC 2020), 2020, : 4095 - 4101
[32] Deep Reinforcement Learning to Assist Command and Control
Park, Song Jun
Vindiola, Manuel M.
Logie, Anne C.
Narayanan, Priya
Davies, Jared
ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING FOR MULTI-DOMAIN OPERATIONS APPLICATIONS IV, 2022, 12113
[33] Deep reinforcement learning for quantum gate control
An, Zheng
Zhou, D. L.
EPL, 2019, 126 (06)
[34] DEEP REINFORCEMENT LEARNING FOR TRANSFER OF CONTROL POLICIES
Cunningham, James D.
Miller, Simon W.
Yukish, Michael A.
Simpson, Timothy W.
Tucker, Conrad S.
PROCEEDINGS OF THE ASME INTERNATIONAL DESIGN ENGINEERING TECHNICAL CONFERENCES AND COMPUTERS AND INFORMATION IN ENGINEERING CONFERENCE, 2019, VOL 2A, 2020,
[35] Control of chaotic systems by deep reinforcement learning
Bucci, M. A.
Semeraro, O.
Allauzen, A.
Wisniewski, G.
Cordier, L.
Mathelin, L.
PROCEEDINGS OF THE ROYAL SOCIETY A-MATHEMATICAL PHYSICAL AND ENGINEERING SCIENCES, 2019, 475 (2231):
[36] Deep Decentralized Reinforcement Learning for Cooperative Control
Koepf, Florian
Tesfazgi, Samuel
Flad, Michael
Hohmann, Soeren
IFAC PAPERSONLINE, 2020, 53 (02): : 1555 - 1562
[37] Deep reinforcement learning control for non-stationary building energy management
Naug, Avisek
Quinones-Grueiro, Marcos
Biswas, Gautam
ENERGY AND BUILDINGS, 2022, 277
[38] Learning to Control Random Boolean Networks: A Deep Reinforcement Learning Approach
Papagiannis, Georgios
Moschoyiannis, Sotiris
COMPLEX NETWORKS AND THEIR APPLICATIONS VIII, VOL 1, 2020, 881 : 721 - 734
[39] Jointly Learning to Construct and Control Agents using Deep Reinforcement Learning
Schaff, Charles
Yunis, David
Chakrabarti, Ayan
Walter, Matthew R.
2019 INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), 2019, : 9798 - 9805
[40] A Deep Reinforcement Learning Approach to Improve the Learning Performance in Process Control
Bao, Yaoyao
Zhu, Yuanming
Qian, Feng
INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH, 2021, 60 (15) : 5504 - 5515

← 1 2 3 4 5 →