Open main menu

Timelines β

Difference between revisions of "Timeline of decision theory"

(What the timeline is still missing)
(Numerical and visual data)
 
(44 intermediate revisions by 3 users not shown)
Line 1: Line 1:
This is a '''timeline of decision theory''', with a focus on updateless/timeless/acausal decision theories.
+
This is a '''timeline of decision theory''', with a focus on updateless/timeless/acausal/functional/logical decision theories.
  
 
==Big picture==
 
==Big picture==
Line 5: Line 5:
 
{| class="wikitable"
 
{| class="wikitable"
 
! Time period !! Development summary !! More details
 
! Time period !! Development summary !! More details
 +
|-
 +
| up to 2006 || the bad old days ||
 +
|-
 +
| 2006-2010 || Drescher, TDT, UDT ||
 +
|-
 +
| 2010-2012 || more progress ||
 +
|-
 +
| 2012 || stagnation? ||
 +
|}
 +
 +
== Numerical and visual data  ==
 +
 +
=== Google Scholar ===
 +
 +
The following table summarizes per-year mentions on Google Scholar as of May 22, 2021.
 +
 +
{| class="sortable wikitable"
 +
! Year
 +
! decision theory
 +
! behavioral decision theory
 +
! bayesian decision theory
 +
! statistical decision theory
 +
! normative decision theory
 +
! descriptive decision theory
 +
|-
 +
| 1980 || 50,700 || 9,980 || 1,170 || 15,500 || 5,060 || 8,280
 +
|-
 +
| 1985 || 69,100 || 11,100 || 1,260 || 17,700 || 5,850 || 8,610
 +
|-
 +
| 1990 || 121,000 || 17,900 || 2,750 || 29,800 || 9,640 || 16,200
 +
|-
 +
| 1995 || 208,000 || 22,700 || 3,870 || 50,300 || 15,700 || 23,100
 +
|-
 +
| 2000 || 386,000 || 43,200 || 6,610 || 102,000 || 32,100 || 43,000
 +
|-
 +
| 2002 || 474,000 || 61,600 || 8,830 || 122,000 || 40,600 || 53,600
 +
|-
 +
| 2004 || 574,000 || 76,100 || 11,900 || 135,000 || 44,700 || 58,200
 +
|-
 +
| 2006 || 727,000 || 90,200 || 15,800 || 161,000 || 62,100 || 76,800
 +
|-
 +
| 2008 || 889,000 || 125,000 || 21,000 || 187,000 || 72,600 || 93,200
 +
|-
 +
| 2010 || 1,020,000 || 144,000 || 26,300 || 211,000 || 82,000 || 110,000
 +
|-
 +
| 2012 || 1,130,000 || 170,000 || 34,000 || 240,000 || 91,800 || 128,000
 +
|-
 +
| 2014 || 993,000 || 177,000 || 38,400 || 222,000 || 89,600 || 126,000
 +
|-
 +
| 2016 || 786,000 || 147,000 || 41,700 || 179,000 || 86,300 || 112,000
 +
|-
 +
| 2017 || 682,000 || 126,000 || 36,200 || 155,000 || 80,500 || 102,000
 +
|-
 +
| 2018 || 465,000 || 116,000 || 45,200 || 129,000 || 69,700 || 78,400
 +
|-
 +
| 2019 || 300,000 || 87,900 || 40,000 || 96,800 || 55,900 || 68.700
 +
|-
 +
| 2020 || 207,000 || 63,500 || 36,400 || 70,800 || 43,800 || 51,400 
 
|-
 
|-
 
|}
 
|}
 +
 +
[[File:Deision theory tb.png|thumb|center|700px]]
 +
 +
=== Google Trends ===
 +
 +
The comparative chart below shows {{w|Google Trends}} data for Decision theory (Field of study) and Decision theory (Search Term) from January 2004 to February 2021, when the screenshot was taken. Interest is also ranked by country and displayed on world map.<ref>{{cite web |title=Decision theory |url=https://trends.google.com/trends/explore?date=all&q=%2Fm%2F029frg,Decision%20theory |website=Google Trends |access-date=15 February 2021}}</ref>
 +
 +
[[File:Decision theory gt.jpeg|thumb|center|600px]]
 +
 +
=== Google Ngram Viewer ===
 +
 +
The chart below shows {{w|Google Ngram Viewer}} data for Decision theory, from 1900 to 2019.<ref>{{cite web |title=Decision theory |url=https://books.google.com/ngrams/graph?content=Decision+theory&year_start=1900&year_end=2019&corpus=26&smoothing=3&direct_url=t1%3B%2CDecision%20theory%3B%2Cc0#t1%3B%2CDecision%20theory%3B%2Cc0 |website=books.google.com |access-date=2 August 2021 |language=en}}</ref>
 +
 +
[[File:Decision theory ngram.png|thumb|center|700px]]
 +
 +
=== Wikipedia Views ===
 +
 +
The chart below shows pageviews of the English Wikipedia article {{w|Decision theory}}, from July 2015 to July 2021.<ref>{{cite web |title=Decision theory |url=https://wikipediaviews.org/displayviewsformultiplemonths.php?page=Decision+theory&allmonths=allmonths-api&language=en&drilldown=all |website=wikipediaviews.org |access-date=2 August 2021}}</ref>
 +
 +
[[File:Decision theory wv.png|thumb|center|550px]]
  
 
==Full timeline==
 
==Full timeline==
Line 22: Line 100:
 
|-
 
|-
 
| 1997 || || || The absent-minded driver problem is introduced (in the same paper as the sleeping beauty?).<ref>{{cite web |url=http://lesswrong.com/lw/182/the_absentminded_driver/ |title=The Absent-Minded Driver |date=September 16, 2009 |accessdate=September 10, 2017 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref><ref>{{cite web |url=https://wiki.lesswrong.com/wiki/Absentminded_driver |title=Absent-Minded driver - Lesswrongwiki |accessdate=September 10, 2017 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref>
 
| 1997 || || || The absent-minded driver problem is introduced (in the same paper as the sleeping beauty?).<ref>{{cite web |url=http://lesswrong.com/lw/182/the_absentminded_driver/ |title=The Absent-Minded Driver |date=September 16, 2009 |accessdate=September 10, 2017 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref><ref>{{cite web |url=https://wiki.lesswrong.com/wiki/Absentminded_driver |title=Absent-Minded driver - Lesswrongwiki |accessdate=September 10, 2017 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref>
 +
|-
 +
| 1999 || {{dts|January 21}} || || Wei Dai posts the first description of what would later be called UDASSA is posted to everything-list.<ref>{{cite web |url=https://riceissa.github.io/everything-list-1998-2009/0316.html |title=Re: consciousness based on information or computation? |date=January 21, 1999 |accessdate=March 6, 2020 |publisher=everything-list |author=Wei Dai}}</ref> UDASSA seems to be a precursor to UDT.<ref>https://www.greaterwrong.com/posts/SkXLrDXyHeekqgbFg/shock-level-5-big-worlds-and-modal-realism/comment/yMCxvHCpBqsYEorpt</ref>
 
|-
 
|-
 
| 2002 || {{dts|July 17}} || || [[wikipedia:Hal Finney (computer scientist)|Hal Finney]], in a mailing list discussion, brings up ideas that according to Wei Dai come "pretty close to some of the ideas behind TDT".<ref>{{cite web |url=http://lesswrong.com/lw/b7v/common_mistakes_people_make_when_thinking_about/65zv |title=Wei_Dai comments on Common mistakes people make when thinking about decision theory - Less Wrong |accessdate=September 10, 2017 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref><ref>{{cite web |url=https://groups.google.com/forum/#!msg/everything-list/V0BSqfkwbLo/YcIaaMYs7w0J |title=self-sampling assumption is incorrect |publisher=Google Groups |date=July 17, 2002 |first=Hal |last=Finney |accessdate=September 10, 2017}}</ref>
 
| 2002 || {{dts|July 17}} || || [[wikipedia:Hal Finney (computer scientist)|Hal Finney]], in a mailing list discussion, brings up ideas that according to Wei Dai come "pretty close to some of the ideas behind TDT".<ref>{{cite web |url=http://lesswrong.com/lw/b7v/common_mistakes_people_make_when_thinking_about/65zv |title=Wei_Dai comments on Common mistakes people make when thinking about decision theory - Less Wrong |accessdate=September 10, 2017 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref><ref>{{cite web |url=https://groups.google.com/forum/#!msg/everything-list/V0BSqfkwbLo/YcIaaMYs7w0J |title=self-sampling assumption is incorrect |publisher=Google Groups |date=July 17, 2002 |first=Hal |last=Finney |accessdate=September 10, 2017}}</ref>
Line 37: Line 117:
 
| 2009 || {{dts|March 19}} || || Vladimir Nesov introduces counterfactual mugging.<ref>{{cite web |url=http://lesswrong.com/lw/3l/counterfactual_mugging/ |title=Counterfactual Mugging |date=March 19, 2009 |first=Vladimir |last=Nesov |accessdate=September 10, 2017 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref><ref>{{cite web |url=https://wiki.lesswrong.com/wiki/Counterfactual_mugging |title=Counterfactual mugging - Lesswrongwiki |accessdate=September 10, 2017 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref>
 
| 2009 || {{dts|March 19}} || || Vladimir Nesov introduces counterfactual mugging.<ref>{{cite web |url=http://lesswrong.com/lw/3l/counterfactual_mugging/ |title=Counterfactual Mugging |date=March 19, 2009 |first=Vladimir |last=Nesov |accessdate=September 10, 2017 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref><ref>{{cite web |url=https://wiki.lesswrong.com/wiki/Counterfactual_mugging |title=Counterfactual mugging - Lesswrongwiki |accessdate=September 10, 2017 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref>
 
|-
 
|-
| 2009 || {{dts|August 13}} || || Wei Dai publishes the post "Towards a New Decision Theory" on LessWrong. The post does not use the term "updateless decision theory" (UDT), but describes what would later be known as UDT.<ref>{{cite web |url=http://lesswrong.com/lw/15m/towards_a_new_decision_theory/ |title=Towards a New Decision Theory - Less Wrong |accessdate=January 10, 2018 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref><ref name="taking_ideas_seriously_comment">{{cite web |url=http://lesswrong.com/lw/2l6/taking_ideas_seriously/2fon |title=Wei_Dai comments on Taking Ideas Seriously - Less Wrong |accessdate=January 10, 2018 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref>
+
| 2009 || {{dts|August 13}} || || Wei Dai publishes the post "Towards a New Decision Theory" on LessWrong. The post does not use the term "updateless decision theory" (UDT), but describes what would later be known as UDT1.<ref>{{cite web |url=http://lesswrong.com/lw/15m/towards_a_new_decision_theory/ |title=Towards a New Decision Theory - Less Wrong |accessdate=January 10, 2018 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref><ref name="taking_ideas_seriously_comment">{{cite web |url=http://lesswrong.com/lw/2l6/taking_ideas_seriously/2fon |title=Wei_Dai comments on Taking Ideas Seriously - Less Wrong |accessdate=January 10, 2018 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref>
 
|-
 
|-
 
| 2009 || {{dts|August 20}} || || {{w|Gary Drescher}} proposes Metacircular Decision Theory (MCDT) in a comment on LessWrong.<ref>{{cite web |url=http://lesswrong.com/lw/15z/ingredients_of_timeless_decision_theory/1217 |title=Gary_Drescher comments on Ingredients of Timeless Decision Theory - Less Wrong |accessdate=September 10, 2017 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref>
 
| 2009 || {{dts|August 20}} || || {{w|Gary Drescher}} proposes Metacircular Decision Theory (MCDT) in a comment on LessWrong.<ref>{{cite web |url=http://lesswrong.com/lw/15z/ingredients_of_timeless_decision_theory/1217 |title=Gary_Drescher comments on Ingredients of Timeless Decision Theory - Less Wrong |accessdate=September 10, 2017 |publisher=[[wikipedia:LessWrong|LessWrong]]}}</ref>
 
|-
 
|-
 
| 2010 || || || Timeless decision theory is published in paper form by {{w|Eliezer Yudkowsky}}.<ref>{{cite web |url=https://intelligence.org/files/TDT.pdf |title=Timeless Decision Theory |first=Eliezer |last=Yudkowsky |year=2010 |accessdate=September 10, 2017}}</ref>
 
| 2010 || || || Timeless decision theory is published in paper form by {{w|Eliezer Yudkowsky}}.<ref>{{cite web |url=https://intelligence.org/files/TDT.pdf |title=Timeless Decision Theory |first=Eliezer |last=Yudkowsky |year=2010 |accessdate=September 10, 2017}}</ref>
 +
|-
 +
| 2010 || {{dts|February 18}} || || Wei Dai publishes "Explicit Optimization of Global Strategy (Fixing a Bug in UDT1)" on LessWrong.<ref>{{cite web |url=https://www.lesswrong.com/posts/g8xh9R7RaNitKtkaa/explicit-optimization-of-global-strategy-fixing-a-bug-in |title=Explicit Optimization of Global Strategy (Fixing a Bug in UDT1) |accessdate=July 25, 2018 |first=Wei |last=Dai |publisher=LessWrong |date=February 18, 2010}}</ref> This post introduces the decision theory UDT1.1, which improves on UDT1 by iterating over policies (observations-to-actions mappings) rather than iterating over actions.
 +
|-
 +
| 2010 || {{dts|April}} || || Gary Drescher proposes the "agent simulates predictor" decision problem to the decision-theory-workshop mailing list (a private mailing list for discussing decision theory).<ref>{{cite web |url=https://www.lesswrong.com/posts/q9DbfYfFzkotno9hG/example-decision-theory-problem-agent-simulates-predictor |title=Example decision theory problem: "Agent simulates predictor" |accessdate=July 25, 2018 |publisher=[[wikipedia:LessWrong|LessWrong]] |date=May 19, 2011 |first=Vladimir |last=Slepnev}}</ref> The problem would be published publicly by Vladimir Slepnev in May 2011.
 +
|-
 +
| 2011 || || || Wei Dai proposes UDT2 in a post to the decision theory workshop mailing list.<ref>{{cite web |url=https://agentfoundations.org/item?id=1078 |website=Intelligent Agent Foundations Forum |publisher=Machine Intelligence Research Institute |title=Comment on "Updatelessness and Son of X" |date=November 6, 2016 |accessdate=July 26, 2018 |quote=This does seem to be the “obvious” next step in the UDT approach. I proposed something similar as “UDT2” in a 2011 post to the “decision theory workshop” mailing list, and others have made similar proposals.}}</ref> The idea behind UDT2 would be described in a comment by Wei Dai in January 2014,<ref>{{cite web |url=https://www.lesswrong.com/posts/WjXYsSqdwTY9sJ5mP/functional-side-effects#mFTEbpgufwyc7SMq2 |date=January 15, 2014 |first=Wei |last=Dai |publisher=LessWrong |title=Comment on "Functional Side Effects" |accessdate=July 26, 2018}}</ref> and by Vladimir Slepnev in a blog post in September 2013.<ref>{{cite web |url=https://www.lesswrong.com/posts/wXxPmc9W6kPb6i7vj/notes-on-logical-priors-from-the-miri-workshop |title=Notes on logical priors from the MIRI workshop |accessdate=July 26, 2018 |publisher=[[wikipedia:LessWrong|LessWrong]] |date=September 15, 2013 |first=Vladimir |last=Slepnev}}</ref>
 +
|-
 +
| 2014 || {{dts|April 23}} || || Daniel Hintze publishes "Problem Class Dominance in Predictive Dilemmas".<ref>{{cite web |url=https://intelligence.org/wp-content/uploads/2014/10/Hintze-Problem-Class-Dominance-In-Predictive-Dilemmas.pdf |title=Problem Class Dominance in Predictive Dilemmas |date=April 23, 2014 |first=Daniel |last=Hintze |publisher=Machine Intelligence Research Institute}}</ref> The paper compares evidential decision theory, causal decision theory, timeless decision theory, and updateless decision theory (specifically, UDT1.1) on the decision problems Parfit's hitchhiker and the curious benefactor (equivalent to counterfactual mugging?).
 +
|-
 +
| 2014 || {{dts|November 4}} || Project || The Intelligent Agent Foundations Forum, run by MIRI, is launched.<ref>{{cite web |url=https://agentfoundations.org/item?id=1 |website=Intelligent Agent Foundations Forum |title=Welcome! |author=Benja Fallenstein |accessdate=June 30, 2017 |quote=post by Benja Fallenstein 969 days ago}}</ref>
 
|-
 
|-
 
| 2017 || {{dts|March 18}} || || "Cheating Death in Damascus" by Nate Soares and Ben Levinstein is announced on the {{w|Machine Intelligence Research Institute}} blog.<ref>{{cite web |url=https://intelligence.org/2017/03/18/new-paper-cheating-death-in-damascus/ |title=New paper: "Cheating Death in Damascus" |first=Rob |last=Bensinger |publisher=Machine Intelligence Research Institute |date=March 18, 2017 |accessdate=September 10, 2017}}</ref><ref>{{cite web |url=https://intelligence.org/files/DeathInDamascus.pdf |title=Cheating Death in Damascus |first1=Nate |last1=Soares |first2=Benjamin A. |last2=Levinstein |accessdate=September 10, 2017}}</ref>
 
| 2017 || {{dts|March 18}} || || "Cheating Death in Damascus" by Nate Soares and Ben Levinstein is announced on the {{w|Machine Intelligence Research Institute}} blog.<ref>{{cite web |url=https://intelligence.org/2017/03/18/new-paper-cheating-death-in-damascus/ |title=New paper: "Cheating Death in Damascus" |first=Rob |last=Bensinger |publisher=Machine Intelligence Research Institute |date=March 18, 2017 |accessdate=September 10, 2017}}</ref><ref>{{cite web |url=https://intelligence.org/files/DeathInDamascus.pdf |title=Cheating Death in Damascus |first1=Nate |last1=Soares |first2=Benjamin A. |last2=Levinstein |accessdate=September 10, 2017}}</ref>
 
|-
 
|-
 
| 2017 || {{dts|October 13}} || || "Functional Decision Theory: A New Theory of Instrumental Rationality" by {{W|Eliezer Yudkowsky}} and Nate Soares is posted to the {{w|arXiv}}.<ref name="fdt">{{cite web |url=https://arxiv.org/abs/1710.05060 |title=[1710.05060] Functional Decision Theory: A New Theory of Instrumental Rationality |accessdate=October 22, 2017 |quote=Submitted on 13 Oct 2017 |first1=Eliezer |last1=Yudkowsky |first2=Nate |last2=Soares}}</ref> The paper is announced on the {{w|Machine Intelligence Research Institute}} blog on October 22.<ref>{{cite web |url=https://intelligence.org/2017/10/22/fdt/ |title=New paper: "Functional Decision Theory" - Machine Intelligence Research Institute |publisher=Machine Intelligence Research Institute |date=October 22, 2017 |author=Matthew Graves |accessdate=October 22, 2017}}</ref>
 
| 2017 || {{dts|October 13}} || || "Functional Decision Theory: A New Theory of Instrumental Rationality" by {{W|Eliezer Yudkowsky}} and Nate Soares is posted to the {{w|arXiv}}.<ref name="fdt">{{cite web |url=https://arxiv.org/abs/1710.05060 |title=[1710.05060] Functional Decision Theory: A New Theory of Instrumental Rationality |accessdate=October 22, 2017 |quote=Submitted on 13 Oct 2017 |first1=Eliezer |last1=Yudkowsky |first2=Nate |last2=Soares}}</ref> The paper is announced on the {{w|Machine Intelligence Research Institute}} blog on October 22.<ref>{{cite web |url=https://intelligence.org/2017/10/22/fdt/ |title=New paper: "Functional Decision Theory" - Machine Intelligence Research Institute |publisher=Machine Intelligence Research Institute |date=October 22, 2017 |author=Matthew Graves |accessdate=October 22, 2017}}</ref>
 +
|-
 +
| 2018 || {{dts|July 10}} || || The Alignment Forum beta is announced.<ref>{{cite web |url=https://www.lesswrong.com/posts/JiMAMNAb55Qq24nES/announcing-alignmentforum-org-beta |title=Announcing AlignmentForum.org Beta |accessdate=July 25, 2018 |publisher=[[wikipedia:LessWrong|LessWrong]] |date=July 10, 2018 |author=Raemon}}</ref> The forum is a website intended for discussing research in AI alignment. (Decision theory is sometimes motivated by AI alignment concerns.)
 
|}
 
|}
  
Line 58: Line 150:
 
===What the timeline is still missing===
 
===What the timeline is still missing===
  
* MIRI agent foundations forum
+
* History of the concept of decision theory
 +
* More on decision theory in academia, journals related to it, where it fits in with the rest of academia
 +
* [https://philpapers.org/rec/CAMTPD symmetry argument]? I found this paper linked in [https://qchu.wordpress.com/2013/06/30/cantors-theorem-the-prisoners-dilemma-and-the-halting-problem/]
 
* https://wiki.lesswrong.com/wiki/Parfit%27s_hitchhiker https://arbital.com/p/parfits_hitchhiker/
 
* https://wiki.lesswrong.com/wiki/Parfit%27s_hitchhiker https://arbital.com/p/parfits_hitchhiker/
* http://fennetic.net/irc/finney.org/~hal/udassa/summary1.html
+
* http://fennetic.net/irc/finney.org/~hal/udassa/summary1.html -- in a post on everything-list, hal also mentions that he at one point had udassa.com. it looks like wayback didn't capture it in time (by the time it got to it, the domain was parked and for sale); i'm not sure if udassa.com had anything different from finney.org/~hal/udassa
 
* http://lesswrong.com/lw/gu1/decision_theory_faq/
 
* http://lesswrong.com/lw/gu1/decision_theory_faq/
 
* http://lesswrong.com/lw/aq9/decision_theories_a_less_wrong_primer/
 
* http://lesswrong.com/lw/aq9/decision_theories_a_less_wrong_primer/
Line 67: Line 161:
 
* Something about Spohn; see e.g. [http://lesswrong.com/lw/eik/eliezers_sequences_and_mainstream_academia/7fx4 this comment]
 
* Something about Spohn; see e.g. [http://lesswrong.com/lw/eik/eliezers_sequences_and_mainstream_academia/7fx4 this comment]
 
* "This idea follows in the wake of Gauthier (1994), who advocated making decisions using global policy selection, and Arntzenius, Elga, and Hawthorne (2004), who applied this idea to an infinite decision problem similar to the “Procrastination Paradox” of Yudkowsky (2013). Another decision procedure similar to that of Dai was proposed by Meacham (2010)" [https://arxiv.org/pdf/1507.01986.pdf]
 
* "This idea follows in the wake of Gauthier (1994), who advocated making decisions using global policy selection, and Arntzenius, Elga, and Hawthorne (2004), who applied this idea to an infinite decision problem similar to the “Procrastination Paradox” of Yudkowsky (2013). Another decision procedure similar to that of Dai was proposed by Meacham (2010)" [https://arxiv.org/pdf/1507.01986.pdf]
 +
* when was TDT "officially" declared obsolete?
 +
* I think several of cousin_it's posts should be included. i don't know enough yet to know which ones though.
 +
** there was one i think that introduced the idea of "playing chicken with the universe"
 +
* "the time they tried to hire a philosophy prof to write up TDT?" [http://nostalgebraist.tumblr.com/post/168738981769/youzicha-nostalgebraist-shlevy] See [https://www.reddit.com/r/askphilosophy/comments/7qvsod/what_do_actual_decision_theorists_think_of/dsu3otc/], [http://lesswrong.com/lw/cok/funding_good_research/], [http://lesswrong.com/lw/f5b/the_problem_with_rational_wiki/7pj4], [https://www.reddit.com/r/askphilosophy/comments/7qvsod/what_do_actual_decision_theorists_think_of/]
 +
* Will MacAskill has a Meta Decision Theory that's supposed to take into account uncertainty about which decision theory to use (sounds kinda similar to his approach to moral uncertainty?) [http://lesswrong.com/r/lesswrong/lw/gw5/meta_decision_theory_and_newcombs_problem/]
 +
* https://news.ycombinator.com/item?id=9321984
 +
* UDT1.5/UDT2
 +
* roko's basilisk
 +
* interesting historical bit about UDT and two-boxing on newcomb: https://www.greaterwrong.com/posts/Kr76XzME7TFkN937z/predictors-exist-cdt-going-bonkers-forever/comment/afyRSrYtx8nP6kCs3 (i think wei dai has an older comment on LW saying a similar thing but giving less detail)
 +
* vanessa kosoy and abram demski's discussion.
 +
* some of jessica taylor's work
 +
* progress or lack thereof for making decision theory work with logical inductors
 +
* macaskill's criticism of FDT (posted on LW)
 +
* big discussion of FDT on buck's AMA on EA forum https://ea.greaterwrong.com/posts/tDk57GhrdK54TWzPY/i-m-buck-shlegeris-i-do-research-and-outreach-at-miri-ama/comment/byH8abnt5RnPMunts
 +
* 5 and 10 problem
 +
* [https://www.greaterwrong.com/posts/h9qQQA3g8dwq6RRTo/counterfactual-mugging-why-should-you-pay/answer/5satru2L8dbyp9Z4s] Abram Demski on the differences between UDT, TDT, and FDT
  
 
===Timeline update strategy===
 
===Timeline update strategy===
Line 74: Line 184:
 
* [[Timeline of Machine Intelligence Research Institute]]
 
* [[Timeline of Machine Intelligence Research Institute]]
 
* [[Timeline of Center for Applied Rationality]]
 
* [[Timeline of Center for Applied Rationality]]
* [[Timeline of friendly artificial intelligence]]
+
* [[Timeline of AI safety]]
 
* [[Timeline of Wei Dai publications]]
 
* [[Timeline of Wei Dai publications]]
  
Line 80: Line 190:
  
 
* [https://casparoesterheld.com/a-comprehensive-list-of-decision-theories/ "A comprehensive list of decision theories"] by Caspar Oesterheld and Johannes Treutlein
 
* [https://casparoesterheld.com/a-comprehensive-list-of-decision-theories/ "A comprehensive list of decision theories"] by Caspar Oesterheld and Johannes Treutlein
 +
* [https://www.alignmentforum.org/posts/QPhY8Nb7gtT5wvoPH/comparison-of-decision-theories-with-a-focus-on-logical "Comparison of decision theories (with a focus on logical-counterfactual decision theories)"] by Issa Rice
  
 
==References==
 
==References==
  
 
{{Reflist|30em}}
 
{{Reflist|30em}}

Latest revision as of 16:10, 2 August 2021

This is a timeline of decision theory, with a focus on updateless/timeless/acausal/functional/logical decision theories.

Contents

Big picture

Time period Development summary More details
up to 2006 the bad old days
2006-2010 Drescher, TDT, UDT
2010-2012 more progress
2012 stagnation?

Numerical and visual data

Google Scholar

The following table summarizes per-year mentions on Google Scholar as of May 22, 2021.

Year decision theory behavioral decision theory bayesian decision theory statistical decision theory normative decision theory descriptive decision theory
1980 50,700 9,980 1,170 15,500 5,060 8,280
1985 69,100 11,100 1,260 17,700 5,850 8,610
1990 121,000 17,900 2,750 29,800 9,640 16,200
1995 208,000 22,700 3,870 50,300 15,700 23,100
2000 386,000 43,200 6,610 102,000 32,100 43,000
2002 474,000 61,600 8,830 122,000 40,600 53,600
2004 574,000 76,100 11,900 135,000 44,700 58,200
2006 727,000 90,200 15,800 161,000 62,100 76,800
2008 889,000 125,000 21,000 187,000 72,600 93,200
2010 1,020,000 144,000 26,300 211,000 82,000 110,000
2012 1,130,000 170,000 34,000 240,000 91,800 128,000
2014 993,000 177,000 38,400 222,000 89,600 126,000
2016 786,000 147,000 41,700 179,000 86,300 112,000
2017 682,000 126,000 36,200 155,000 80,500 102,000
2018 465,000 116,000 45,200 129,000 69,700 78,400
2019 300,000 87,900 40,000 96,800 55,900 68.700
2020 207,000 63,500 36,400 70,800 43,800 51,400
Deision theory tb.png

Google Trends

The comparative chart below shows Google Trends data for Decision theory (Field of study) and Decision theory (Search Term) from January 2004 to February 2021, when the screenshot was taken. Interest is also ranked by country and displayed on world map.[1]

Decision theory gt.jpeg

Google Ngram Viewer

The chart below shows Google Ngram Viewer data for Decision theory, from 1900 to 2019.[2]

Decision theory ngram.png

Wikipedia Views

The chart below shows pageviews of the English Wikipedia article Decision theory, from July 2015 to July 2021.[3]

Decision theory wv.png

Full timeline

Year Month and date Event type Details
1969 Newcomb's problem is discussed by Robert Nozick.
1980 Brian Skyrms's Causal Necessity: A Pragmatic Investigation of the Necessity of Laws discusses the smoking lesion problem (or a similar problem that becomes called the smoking lesion problem in later publications).[4]:128–130 Yudkowsky and Soares cite Skyrms for the smoking lesion problem.[5]:3
1985 The idea of superrationality is introduced by Douglas Hofstadter in his Metamagical Themas.
1997 The Sleeping Beauty problem is first formally analyzed.
1997 The absent-minded driver problem is introduced (in the same paper as the sleeping beauty?).[6][7]
1999 January 21 Wei Dai posts the first description of what would later be called UDASSA is posted to everything-list.[8] UDASSA seems to be a precursor to UDT.[9]
2002 July 17 Hal Finney, in a mailing list discussion, brings up ideas that according to Wei Dai come "pretty close to some of the ideas behind TDT".[10][11]
2006 March 29 On the Theory of Everything Mailing List (everything-list), Wei Dai sends an email with subject "proper behavior for a mathematical substructure". He would later call this "a 2006 proto-UDT".[12]
2006 May 5 Gary Drescher's Good and Real is published.[13]
2007 The Smoking Lesion problem is introduced by Andy Egan?[14][15]
2007 May 30 Philosopher Kenny Easwaran blogs about his discussions with Joshua Von Korff. Korff has apparently devised a decision-theoretic protocol that one-boxes on Newcomb's problem but smokes in the Smoking Lesion problem. The post does not make clear when Korff came up with his ideas or whether he wrote them up anywhere.[16][17]
2009 February Eliezer Yudkowsky starts LessWrong using as seed material his posts on Overcoming Bias.[18] During the following years LessWrong would become the locus of discussion about timeless/updateless decision theory.
2009 March 19 Vladimir Nesov introduces counterfactual mugging.[19][20]
2009 August 13 Wei Dai publishes the post "Towards a New Decision Theory" on LessWrong. The post does not use the term "updateless decision theory" (UDT), but describes what would later be known as UDT1.[21][12]
2009 August 20 Gary Drescher proposes Metacircular Decision Theory (MCDT) in a comment on LessWrong.[22]
2010 Timeless decision theory is published in paper form by Eliezer Yudkowsky.[23]
2010 February 18 Wei Dai publishes "Explicit Optimization of Global Strategy (Fixing a Bug in UDT1)" on LessWrong.[24] This post introduces the decision theory UDT1.1, which improves on UDT1 by iterating over policies (observations-to-actions mappings) rather than iterating over actions.
2010 April Gary Drescher proposes the "agent simulates predictor" decision problem to the decision-theory-workshop mailing list (a private mailing list for discussing decision theory).[25] The problem would be published publicly by Vladimir Slepnev in May 2011.
2011 Wei Dai proposes UDT2 in a post to the decision theory workshop mailing list.[26] The idea behind UDT2 would be described in a comment by Wei Dai in January 2014,[27] and by Vladimir Slepnev in a blog post in September 2013.[28]
2014 April 23 Daniel Hintze publishes "Problem Class Dominance in Predictive Dilemmas".[29] The paper compares evidential decision theory, causal decision theory, timeless decision theory, and updateless decision theory (specifically, UDT1.1) on the decision problems Parfit's hitchhiker and the curious benefactor (equivalent to counterfactual mugging?).
2014 November 4 Project The Intelligent Agent Foundations Forum, run by MIRI, is launched.[30]
2017 March 18 "Cheating Death in Damascus" by Nate Soares and Ben Levinstein is announced on the Machine Intelligence Research Institute blog.[31][32]
2017 October 13 "Functional Decision Theory: A New Theory of Instrumental Rationality" by Eliezer Yudkowsky and Nate Soares is posted to the arXiv.[5] The paper is announced on the Machine Intelligence Research Institute blog on October 22.[33]
2018 July 10 The Alignment Forum beta is announced.[34] The forum is a website intended for discussing research in AI alignment. (Decision theory is sometimes motivated by AI alignment concerns.)

Meta information on the timeline

How the timeline was built

The initial version of the timeline was written by Issa Rice.

What the timeline is still missing

Timeline update strategy

See also

External links

References

  1. "Decision theory". Google Trends. Retrieved 15 February 2021. 
  2. "Decision theory". books.google.com. Retrieved 2 August 2021. 
  3. "Decision theory". wikipediaviews.org. Retrieved 2 August 2021. 
  4. Skyrms, Brian (1980). Causal Necessity: A Pragmatic Investigation of the Necessity of Laws. Yale University Press. Suppose that the connection between hardening of the arteries and cholesterol intake turned out to be like this: hardening of the arteries is not caused by cholesterol intake like the clogging of a water pipe; rather it is caused by a lesion in the artery wall. In an advanced state these lesions will catch cholesterol from the blood, a fact which has deceived previous researchers about the causal picture. Moreover, imagine that once someone develops the lesion he tends to increase his cholesterol intake. We do not know what mechanism accounts for this effect of the lesion. We do, however, know that the increased cholesterol intake is beneficial; it somehow slows the development of the lesion. Cholesterol intake among those who do not have the lesion appears to have no effect on vascular health. Given this (partly) fanciful account of the etiology of atherosclerosis, what would a rational man who believed the account do when made an offer of Eggs Benedict for breakfast? I say he would accept. He would be a fool to try to "make it the case that he had not developed the lesion" by curtailing his cholesterol intake. […] Examples could be multiplied. R. A. Fisher once suggested that the correlation between smoking and lung cancer might be due to them both being effects of a common genetic cause. Fisher's hypothesis has not fared well, but if, contrary to evidence, it were true and you knew it to be true, and smoking were consistently pleasurable and not harmful in other ways, you would be foolish to refrain from smoking in order to lower the probability of having smoking-cancer gene. You either have it or not, and you can't influence your genetic makeup by abstinence. 
  5. 5.0 5.1 Yudkowsky, Eliezer; Soares, Nate. "[1710.05060] Functional Decision Theory: A New Theory of Instrumental Rationality". Retrieved October 22, 2017. Submitted on 13 Oct 2017 
  6. "The Absent-Minded Driver". LessWrong. September 16, 2009. Retrieved September 10, 2017. 
  7. "Absent-Minded driver - Lesswrongwiki". LessWrong. Retrieved September 10, 2017. 
  8. Wei Dai (January 21, 1999). "Re: consciousness based on information or computation?". everything-list. Retrieved March 6, 2020. 
  9. https://www.greaterwrong.com/posts/SkXLrDXyHeekqgbFg/shock-level-5-big-worlds-and-modal-realism/comment/yMCxvHCpBqsYEorpt
  10. "Wei_Dai comments on Common mistakes people make when thinking about decision theory - Less Wrong". LessWrong. Retrieved September 10, 2017. 
  11. Finney, Hal (July 17, 2002). "self-sampling assumption is incorrect". Google Groups. Retrieved September 10, 2017. 
  12. 12.0 12.1 "Wei_Dai comments on Taking Ideas Seriously - Less Wrong". LessWrong. Retrieved January 10, 2018. 
  13. "Good and Real: Demystifying Paradoxes from Physics to Ethics (MIT Press): Gary L. Drescher: 9780262042338: Amazon.com: Books". Retrieved September 10, 2017. 
  14. "Andy Egan, Some counterexamples to causal decision theory". PhilPapers. Retrieved September 10, 2017. 
  15. "Smoking lesion - Lesswrongwiki". LessWrong. Retrieved September 10, 2017. 
  16. "Different Ideas About Newcomb Cases". Thoughts Arguments and Rants. May 30, 2007. Retrieved September 10, 2017. 
  17. "CarlShulman comments on Counterfactual Mugging". LessWrong. June 21, 2013. Retrieved September 10, 2017. 
  18. "FAQ - Lesswrongwiki". LessWrong. Retrieved June 1, 2017. 
  19. Nesov, Vladimir (March 19, 2009). "Counterfactual Mugging". LessWrong. Retrieved September 10, 2017. 
  20. "Counterfactual mugging - Lesswrongwiki". LessWrong. Retrieved September 10, 2017. 
  21. "Towards a New Decision Theory - Less Wrong". LessWrong. Retrieved January 10, 2018. 
  22. "Gary_Drescher comments on Ingredients of Timeless Decision Theory - Less Wrong". LessWrong. Retrieved September 10, 2017. 
  23. Yudkowsky, Eliezer (2010). "Timeless Decision Theory" (PDF). Retrieved September 10, 2017. 
  24. Dai, Wei (February 18, 2010). "Explicit Optimization of Global Strategy (Fixing a Bug in UDT1)". LessWrong. Retrieved July 25, 2018. 
  25. Slepnev, Vladimir (May 19, 2011). "Example decision theory problem: "Agent simulates predictor"". LessWrong. Retrieved July 25, 2018. 
  26. "Comment on "Updatelessness and Son of X"". Intelligent Agent Foundations Forum. Machine Intelligence Research Institute. November 6, 2016. Retrieved July 26, 2018. This does seem to be the “obvious” next step in the UDT approach. I proposed something similar as “UDT2” in a 2011 post to the “decision theory workshop” mailing list, and others have made similar proposals. 
  27. Dai, Wei (January 15, 2014). "Comment on "Functional Side Effects"". LessWrong. Retrieved July 26, 2018. 
  28. Slepnev, Vladimir (September 15, 2013). "Notes on logical priors from the MIRI workshop". LessWrong. Retrieved July 26, 2018. 
  29. Hintze, Daniel (April 23, 2014). "Problem Class Dominance in Predictive Dilemmas" (PDF). Machine Intelligence Research Institute. 
  30. Benja Fallenstein. "Welcome!". Intelligent Agent Foundations Forum. Retrieved June 30, 2017. post by Benja Fallenstein 969 days ago 
  31. Bensinger, Rob (March 18, 2017). "New paper: "Cheating Death in Damascus"". Machine Intelligence Research Institute. Retrieved September 10, 2017. 
  32. Soares, Nate; Levinstein, Benjamin A. "Cheating Death in Damascus" (PDF). Retrieved September 10, 2017. 
  33. Matthew Graves (October 22, 2017). "New paper: "Functional Decision Theory" - Machine Intelligence Research Institute". Machine Intelligence Research Institute. Retrieved October 22, 2017. 
  34. Raemon (July 10, 2018). "Announcing AlignmentForum.org Beta". LessWrong. Retrieved July 25, 2018.