Skip to main content
23 events
when toggle format what by license comment
Sep 18, 2023 at 2:53 answer added zhixin timeline score: 0
Apr 24, 2023 at 22:46 answer added NoTisan timeline score: 2
Oct 5, 2022 at 10:59 answer added Ashok Chhetri timeline score: 0
May 25, 2022 at 6:00 history tweeted twitter.com/StackStats/status/1529341509233000450
May 24, 2022 at 6:42 answer added Anton timeline score: 1
Feb 7, 2021 at 11:59 comment added Francesco Maybe this could be useful: On-Policy v/s Off-Policy Learning
Jun 14, 2020 at 21:35 answer added Sushil Thapa timeline score: 7
Apr 2, 2020 at 1:36 answer added João Pedro timeline score: 2
Nov 23, 2019 at 4:17 comment added raksheetbhat I found this blog really helpful: leimao.github.io/blog/RL-On-Policy-VS-Off-Policy
Feb 14, 2019 at 9:51 answer added Identicon timeline score: 3
Nov 13, 2018 at 20:50 answer added Dmitry Mottl timeline score: 78
Jun 18, 2018 at 11:27 comment added Josh Albert I would also like to add that there is an off-policy variant of SARSA. This paper (cs.ox.ac.uk/people/shimon.whiteson/pubs/vanseijenadprl09.pdf) will review on and off policy in the introduction, and then explain expected sarsa. Also lookup expected policy gradients (EPG) to find a more general theory that meshes the two types.
Jan 23, 2018 at 20:19 answer added user82135 timeline score: 40
S Dec 17, 2017 at 0:49 history suggested user82135 CC BY-SA 3.0
Polished the question.
Dec 17, 2017 at 0:06 review Suggested edits
S Dec 17, 2017 at 0:49
Aug 27, 2017 at 4:41 answer added Juli timeline score: 11
Jun 20, 2017 at 18:50 comment added Ivan Kush here is a good explanation nb4799.neu.edu/wordpress/?p=1850
S May 16, 2017 at 22:32 history suggested Sobi CC BY-SA 3.0
Part of the question is a quotation from a book. Added quotations to the appropriate part of the question.
May 16, 2017 at 22:04 review Suggested edits
S May 16, 2017 at 22:32
Jan 2, 2017 at 4:27 comment added zyxue I added a comment to stackoverflow.com/questions/6848828/…, the TL;NR part might be helpful with the understanding, too.
Dec 7, 2015 at 4:41 vote accept cgo
Dec 3, 2015 at 8:48 answer added Neil G timeline score: 223
Dec 2, 2015 at 14:21 history asked cgo CC BY-SA 3.0