Value Functions
Value Function is a real-valued function that assigns to each state (or state-action pair) the expected cumulative reward an agent can achieve from that point onward under a given policy, distinguishing it from other solution concepts in Markov Decision Processes by representing long-term desirability rather than immediate reward or action selection.
1.4K
Publications
126.9K
Citations
2.7K
Authors
882
Institutions
Publications per year
2017–2026
485
Parents
1.4K
Deep Reinforcement Learning: A Brief Survey
Kai Arulkumaran, Marc Peter Deisenroth, Miles Brundage et al. · IEEE Signal Processing Magazine · 2017 · 4.3K citations
1–5 of 1.4K
2.7K
18
h-index
17
h-index
15
h-index
15
h-index
13
h-index
1–5 of 2.7K
882

University of Alberta
Edmonton, Canada
24 h-index
84 publications
3.8K citations
Google (United States)
Mountain View, United States
24 h-index
135 publications
7.5K citations
Carnegie Mellon University
Pittsburgh, United States
23 h-index
48 publications
13.9K citations
Stanford University
Stanford, United States
23 h-index
53 publications
5.1K citations
University College London
London, United Kingdom
22 h-index
39 publications
3.7K citations
Google DeepMind (United Kingdom)
London, United Kingdom
21 h-index
82 publications
27.1K citations
Massachusetts Institute of Technology
Cambridge, United States
21 h-index
49 publications
3.8K citations
University of California, Berkeley
Berkeley, United States
20 h-index
46 publications
12.4K citations
1–8 of 882
Repository · 201 publications · 21.2K citations
62
h-index
Conference · 36 publications · 2.4K citations
21
h-index
JAIR · 17 publications · 13.4K citations
16
h-index
15 publications · 1.2K citations
14
h-index
16 publications · 642 citations
14
h-index
1–5