Abstraction for deep reinforcement learning
File(s)Shanahan Mitchell IJCAI 2022.pdf (251.96 KB)
Accepted version
Author(s)
Shanahan, Murray
Mitchell, Melanie
Type
Conference Paper
Abstract
We characterise the problem of abstraction in the
context of deep reinforcement learning. Various
well established approaches to analogical reasoning
and associative memory might be brought to bear
on this issue, but they present difficulties because
of the need for end-to-end differentiability. We re-
view developments in AI and machine learning that
could facilitate their adoption.
context of deep reinforcement learning. Various
well established approaches to analogical reasoning
and associative memory might be brought to bear
on this issue, but they present difficulties because
of the need for end-to-end differentiability. We re-
view developments in AI and machine learning that
could facilitate their adoption.
Date Issued
2022-07-23
Date Acceptance
2022-04-14
Citation
IJCAI : proceedings of the conference / sponsored by the International Joint Conferences on Artificial Intelligence, 2022, pp.5588-5596
ISSN
1045-0823
Publisher
IJCAI
Start Page
5588
End Page
5596
Journal / Book Title
IJCAI : proceedings of the conference / sponsored by the International Joint Conferences on Artificial Intelligence
Copyright Statement
© 2022 The Author(s).
Identifier
https://www.ijcai.org/proceedings/2022/780
Source
International Joint Conference on Artificial Intelligence (IJCAI-ECAI 2022)
Publication Status
Published
Start Date
2022-07-23
Finish Date
2022-07-29
Coverage Spatial
Vienna, Austria
Date Publish Online
2022-07-23