Modified Error Bounds for Matrix Completion and Application to RL

Burnwal, Shantanu Prasad and Vidyasagar, Mathukumalli (2022) Modified Error Bounds for Matrix Completion and Application to RL. IEEE Control Systems Letters, 6. pp. 1916-1921. ISSN 2475-1456

[img] Text
IEEE_Signal_Processing.pdf - Published Version
Available under License Creative Commons Attribution.

Download (828kB)

Abstract

In matrix completion under noisy measurements, most available results assume that there is an a priori bound on the Frobenius norm of the noise, and derive bounds on the Frobenius norm of the residual error. In this letter, we obtain 'component-wise' bounds on the residual error, based on a similar bound on the noise. This among the first such results. As in our earlier paper, choose the locations of the samples to correspond to the edge set of a Ramanujan bigraph. One recent application is to deriving nearly optimal action-value functions in Reinforcement Learning (RL), which is illustrated through examples. The results presented here are only sufficient conditions. Then we illustrate through numerical simulations that there is considerable room for improvement in the sufficient conditions derived here. This is a problem for future research. © 2017 IEEE.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Vidyasagar, MathukumalliUNSPECIFIED
Item Type: Article
Additional Information: This work was supported by the Science and Engineering Research Board (SERB), India.
Uncontrolled Keywords: iterative learning control; machine learning; Statistical learning
Subjects: Electrical Engineering
Divisions: Department of Electrical Engineering
Depositing User: . LibTrainee 2021
Date Deposited: 28 Jul 2022 07:30
Last Modified: 28 Jul 2022 07:30
URI: http://raiith.iith.ac.in/id/eprint/9979
Publisher URL: http://doi.org/10.1109/LCSYS.2021.3133798
OA policy: https://v2.sherpa.ac.uk/id/publication/37771
Related URLs:

    Actions (login required)

    View Item View Item
    Statistics for RAIITH ePrint 9979 Statistics for this ePrint Item