User: Guest  Login
Title:

Analysis and Optimisation of Bellman Residual Errors with Neural Function Approximation

Document type:
Zeitschriftenaufsatz
Author(s):
Gottwald, Martin; Gronauer, Sven; Shen, Hao; Diepold, Klaus
Abstract:
Recent development of Deep Reinforcement Learning (DRL) has demonstrated superior performance of neural networks in solving challenging problems with large or even continuous state spaces. One specific approach is to deploy neural networks to approximate value functions by minimising the Mean Squared Bellman Error (MSBE) function. Despite great successes of DRL, development of reliable and efficient numerical algorithms to minimise the MSBE is still of great scientific interest and practical dem...     »
Keywords:
Critical Point Analysis, Dynamic Programming, Gauss Newton Algorithm, Local Quadratic Convergence, Mean Squared Bellman Error, Residual Gradient
Dewey Decimal Classification:
620 Ingenieurwissenschaften
Journal title:
CoRR
Year:
2021
Journal volume:
abs/2106.08774
Year / month:
2021-10
Quarter:
4. Quartal
Month:
Oct
Reviewed:
nein
Language:
en
WWW:
https://arxiv.org/abs/2106.08774
TUM Institution:
Lehrstuhl für Datenverarbeitung
Last change:
02.08.2022
CC license:
by-nc-sa, http://creativecommons.org/licenses/by-nc-sa/3.0/de
 BibTeX