Cognitive Systems Research, volume 65, pages 74-78

Hierarchical Deep Q-Network from imperfect demonstrations in Minecraft

Publication typeJournal Article
Publication date2021-01-01
Quartile SCImago
Q1
Quartile WOS
Q1
Impact factor3.9
ISSN13890417
Artificial Intelligence
Software
Experimental and Cognitive Psychology
Cognitive Neuroscience
Abstract
We present Hierarchical Deep Q-Network (HDQfD) that won first place in the MineRL competition. The HDQfD works on imperfect demonstrations and utilizes the hierarchical structure of expert trajectories. We introduce the procedure of extracting an effective sequence of meta-actions and subgoals from the demonstration data. We present a structured task-dependent replay buffer and an adaptive prioritizing technique that allow the HDQfD agent to gradually erase poor-quality expert data from the buffer. In this paper, we present the details of the HDQfD algorithm and give the experimental results in the Minecraft domain.

Citations by journals

1
2
3
Lecture Notes in Computer Science
Lecture Notes in Computer Science, 3, 30%
Lecture Notes in Computer Science
3 publications, 30%
Studies in Computational Intelligence
Studies in Computational Intelligence, 1, 10%
Studies in Computational Intelligence
1 publication, 10%
Scientific Programming
Scientific Programming, 1, 10%
Scientific Programming
1 publication, 10%
Computational Intelligence and Neuroscience
Computational Intelligence and Neuroscience, 1, 10%
Computational Intelligence and Neuroscience
1 publication, 10%
Pattern Recognition and Image Analysis
Pattern Recognition and Image Analysis, 1, 10%
Pattern Recognition and Image Analysis
1 publication, 10%
Lecture Notes in Networks and Systems
Lecture Notes in Networks and Systems, 1, 10%
Lecture Notes in Networks and Systems
1 publication, 10%
1
2
3

Citations by publishers

1
2
3
4
5
Springer Nature
Springer Nature, 5, 50%
Springer Nature
5 publications, 50%
Hindawi Limited
Hindawi Limited, 2, 20%
Hindawi Limited
2 publications, 20%
Pleiades Publishing
Pleiades Publishing, 1, 10%
Pleiades Publishing
1 publication, 10%
1
2
3
4
5
  • We do not take into account publications that without a DOI.
  • Statistics recalculated only for publications connected to researchers, organizations and labs registered on the platform.
  • Statistics recalculated weekly.
Metrics
Share
Cite this
GOST |
Cite this
GOST Copy
Skrynnik A. et al. Hierarchical Deep Q-Network from imperfect demonstrations in Minecraft // Cognitive Systems Research. 2021. Vol. 65. pp. 74-78.
GOST all authors (up to 50) Copy
Skrynnik A., Staroverov A., Aitygulov E., Aksenov K., Davydov V., Panov A. Hierarchical Deep Q-Network from imperfect demonstrations in Minecraft // Cognitive Systems Research. 2021. Vol. 65. pp. 74-78.
RIS |
Cite this
RIS Copy
TY - JOUR
DO - 10.1016/j.cogsys.2020.08.012
UR - https://doi.org/10.1016%2Fj.cogsys.2020.08.012
TI - Hierarchical Deep Q-Network from imperfect demonstrations in Minecraft
T2 - Cognitive Systems Research
AU - Skrynnik, Alexey
AU - Staroverov, Aleksei
AU - Aitygulov, Ermek
AU - Aksenov, Kirill
AU - Davydov, Vasilii
AU - Panov, Aleksandr
PY - 2021
DA - 2021/01/01 00:00:00
PB - Elsevier
SP - 74-78
VL - 65
SN - 1389-0417
ER -
BibTex
Cite this
BibTex Copy
@article{2021_Skrynnik,
author = {Alexey Skrynnik and Aleksei Staroverov and Ermek Aitygulov and Kirill Aksenov and Vasilii Davydov and Aleksandr Panov},
title = {Hierarchical Deep Q-Network from imperfect demonstrations in Minecraft},
journal = {Cognitive Systems Research},
year = {2021},
volume = {65},
publisher = {Elsevier},
month = {jan},
url = {https://doi.org/10.1016%2Fj.cogsys.2020.08.012},
pages = {74--78},
doi = {10.1016/j.cogsys.2020.08.012}
}
Found error?