Current browse context:
math.DS
Change to browse by:
References & Citations
Mathematics > Dynamical Systems
Title: Algorithms of Data Development For Deep Learning and Feedback Design
(Submitted on 1 Dec 2019 (v1), last revised 28 Jan 2020 (this version, v2))
Abstract: Recent research reveals that deep learning is an effective way of solving high dimensional Hamilton-Jacobi-Bellman equations. The resulting feedback control law in the form of a neural network is computationally efficient for real-time applications of optimal control. A critical part of this design method is to generate data for training the neural network and validating its accuracy. In this paper, we provide a survey of existing algorithms that can be used to generate data. All the algorithms surveyed in this paper are causality-free, i.e., the solution at a point is computed without using the value of the function at any other points. At the end of the paper, an illustrative example of optimal feedback design using deep learning is given.
Submission history
From: Wei Kang [view email][v1] Sun, 1 Dec 2019 20:12:14 GMT (163kb)
[v2] Tue, 28 Jan 2020 22:25:08 GMT (164kb)
Link back to: arXiv, form interface, contact.