#decisiontheory — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #decisiontheory, aggregated by home.social.
-
We Stopped Using the Mathematics That Works
https://gfrm.in/posts/why-decision-theory-lost/index.html
#HackerNews #WeStoppedUsingMathematics #DecisionTheory #LossOfMathematics #HackerNews #Insights #Innovation
-
We Stopped Using the Mathematics That Works
https://gfrm.in/posts/why-decision-theory-lost/index.html
#HackerNews #WeStoppedUsingMathematics #DecisionTheory #LossOfMathematics #HackerNews #Insights #Innovation
-
Thank you for this fun decision problem!
To start getting a picture I examined the much simpler case of a "two-sided die" or a coin with faces 1 and 2. Same rules otherwise.
Below is a sketch of a truncated decision tree for this case, when the initial roll is "1". Decision nodes are blue squares with decisions as solid blue lines; the decisions are "K" for "keep" and "R" for "roll". Inference/chance nodes are red circles with outcomes as dashed red lines; outcomes are "1" and "2"; it's understood that each has 50% probability.
The expected utility of each decision is in grey at the end of the corresponding line. The fold-back value at each decision branch is in grey above the corresponding decision node.
If the first roll is "2" then it's clearly best to keep it, as the subsequent roll would yield a divisor and a 0$ outcome.
If the first roll is "1", the best decision apparently is to roll once more, and then, if the outcome is "2", keep the result.
Assume that the player stops in any case with "Keep" at the Nth decision, say N = 5. Then the expected utility of the previous "Roll" decision is 9$/2, which is less that the utility of the "Keep" decision, $7. So the previous decision should be "Keep". Folding backwards this situation remains up to the very first decision, for which the expected utility of "Roll" is 3$/2, whereas that for "Keep" is 1$.
It *seems* that the reasoning above would still apply if the player could continue indefinitely, since at each decision the utility of "Keep" is (1+2N)$, whereas that of "Roll" is (1.5 + N)$. But I'm not completely sure about this, there may be some logical gap.
Unfortunately this simplest case is too special, owing to its binary nature, to say something about your score-bound-strategy assumption. It'll be cool to check a 3-sided die :)
-
Thank you for this fun decision problem!
To start getting a picture I examined the much simpler case of a "two-sided die" or a coin with faces 1 and 2. Same rules otherwise.
Below is a sketch of a truncated decision tree for this case, when the initial roll is "1". Decision nodes are blue squares with decisions as solid blue lines; the decisions are "K" for "keep" and "R" for "roll". Inference/chance nodes are red circles with outcomes as dashed red lines; outcomes are "1" and "2"; it's understood that each has 50% probability.
The expected utility of each decision is in grey at the end of the corresponding line. The fold-back value at each decision branch is in grey above the corresponding decision node.
If the first roll is "2" then it's clearly best to keep it, as the subsequent roll would yield a divisor and a 0$ outcome.
If the first roll is "1", the best decision apparently is to roll once more, and then, if the outcome is "2", keep the result.
Assume that the player stops in any case with "Keep" at the Nth decision, say N = 5. Then the expected utility of the previous "Roll" decision is 9$/2, which is less that the utility of the "Keep" decision, $7. So the previous decision should be "Keep". Folding backwards this situation remains up to the very first decision, for which the expected utility of "Roll" is 3$/2, whereas that for "Keep" is 1$.
It *seems* that the reasoning above would still apply if the player could continue indefinitely, since at each decision the utility of "Keep" is (1+2N)$, whereas that of "Roll" is (1.5 + N)$. But I'm not completely sure about this, there may be some logical gap.
Unfortunately this simplest case is too special, owing to its binary nature, to say something about your score-bound-strategy assumption. It'll be cool to check a 3-sided die :)
-
Jon Baron shared Peter Wakker's annotated bibliography of #decisionTheory
> 9000 entries!
DocX http://personal.eur.nl/wakker/refs/webrfrncs.docx
PDF http://personal.eur.nl/wakker/refs/webrfrncs.pdf
BibTeX (no annotations, I merged redundancies) https://www.dropbox.com/scl/fi/vbkki82h62ydq0fol1g8d/Decision-Theory.bib?rlkey=84m8zx3tyaa4uy6p0zptkybnx&st=qx9myebb&dl=0
-
Jon Baron shared Peter Wakker's annotated bibliography of #decisionTheory
> 9000 entries!
DocX http://personal.eur.nl/wakker/refs/webrfrncs.docx
PDF http://personal.eur.nl/wakker/refs/webrfrncs.pdf
BibTeX (no annotations, I merged redundancies) https://www.dropbox.com/scl/fi/vbkki82h62ydq0fol1g8d/Decision-Theory.bib?rlkey=84m8zx3tyaa4uy6p0zptkybnx&st=qx9myebb&dl=0
-
What is #Intelligence?
In a nutshell: If Life is about #survival and #reproduction, Intelligence is the art of making decisions in a complex world that support these goals, through:
🧩 #Modeling – simplify the world
💓 #Feeling – value what matters🔗 https://medium.com/@modfeel/modfeel-coming-soon-in-english-800a209238e5
#modfeel #AI #Philosophy #Cognition #Complexity #DecisionTheory #Evolution #biology #science
-
To be further accurate, you would also not know how many people are on each track, or whether there are any people at all on each track, or how many tracks there are.
#PascalsWager #trolleyProblem #decisionTheory #apologetics
h/t @AnswersInReason
-
Based on work by psychologists Daniel Kahneman and Amos Tversky, who have shown bad feelings about losses are stronger than good feelings we have about gains, Schwartz argues that as you’re presented with countless choices, your pleasure at the prospect of more options is canceled out by the anticipated loss of making a wrong choice.
#Choice #DecisionTheory #dating #optimization #socialnetworking
https://nautil.us/the-problem-with-modern-romance-is-too-much-choice-236135/ -
Based on work by psychologists Daniel Kahneman and Amos Tversky, who have shown bad feelings about losses are stronger than good feelings we have about gains, Schwartz argues that as you’re presented with countless choices, your pleasure at the prospect of more options is canceled out by the anticipated loss of making a wrong choice.
#Choice #DecisionTheory #dating #optimization #socialnetworking
https://nautil.us/the-problem-with-modern-romance-is-too-much-choice-236135/ -
DECISION MAKING ADVICE:
Whenever someone complains that a kid is doing something, respond with, "It would be better if they were doing drugs." because the absurdity of that response is directly proportional to the absurdity of the complaint.
-
DECISION MAKING ADVICE:
Whenever someone complains that a kid is doing something, respond with, "It would be better if they were doing drugs." because the absurdity of that response is directly proportional to the absurdity of the complaint.
-
Been having a bit of existential dread these days. I've got tremendous buckets of skill and talent in building mathematical models and doing decision making calculations and etc, but when I hear about people actually doing that stuff its like "I ran this linear regression and now I don't know how to use the outcomes and what sort of p value can I calculate to tell me if what I did was significant?" Or whatever. The world doesn't WANT competence. #bayesian #decisionTheory #statistics
-
Been having a bit of existential dread these days. I've got tremendous buckets of skill and talent in building mathematical models and doing decision making calculations and etc, but when I hear about people actually doing that stuff its like "I ran this linear regression and now I don't know how to use the outcomes and what sort of p value can I calculate to tell me if what I did was significant?" Or whatever. The world doesn't WANT competence. #bayesian #decisionTheory #statistics
-
from the standpoint of model selection, parsimony often boils down to dimensionality reduction
#modelSelection #parsimony #OccamsRazor #dimensionalityReduction #degreesOfFreedom #complexity #informationTheory #biasVarianceTradeoff #overfitting #underfitting #optimization #parameterTuning #crossValidation #inverseProblems #inference #statisticalLearning #machineLearning #ML #dataScience #modeling #decisionTheory #fitting #regression #classification #residualError #costFunction #performanceLoss
-
from the standpoint of model selection, parsimony often boils down to dimensionality reduction
#modelSelection #parsimony #OccamsRazor #dimensionalityReduction #degreesOfFreedom #complexity #informationTheory #biasVarianceTradeoff #overfitting #underfitting #optimization #parameterTuning #crossValidation #inverseProblems #inference #statisticalLearning #machineLearning #ML #dataScience #modeling #decisionTheory #fitting #regression #classification #residualError #costFunction #performanceLoss
-
A super late #introduction
Hi Mastodon, and thank you @kinozhao for running @fediphilosopher !
I'm Xin Hui, a PhD student in #philosophy shuttling between the University of #Pittsburgh and #MIT. My main philosophical interests lie in the intersection of #decisiontheory and #feminist epistemology, particularly how decision models bear on our #social and #political agency.
You may know me as thyacinth on #twitter!
-
A super late #introduction
Hi Mastodon, and thank you @kinozhao for running @fediphilosopher !
I'm Xin Hui, a PhD student in #philosophy shuttling between the University of #Pittsburgh and #MIT. My main philosophical interests lie in the intersection of #decisiontheory and #feminist epistemology, particularly how decision models bear on our #social and #political agency.
You may know me as thyacinth on #twitter!
-
You have a problem: you currently pick thresholds for model-based actions using some arbitrary heuristic.
Your solution: pick the threshold that maximizes expected utility (e.g. revenue, profit, ROI, …) instead. That’s the definition of the rational decision, right?
Hmm, for some reason you now seem to have several more problems.
#DecisionTheory #Optimization #rationality #AppliedML -
You have a problem: you currently pick thresholds for model-based actions using some arbitrary heuristic.
Your solution: pick the threshold that maximizes expected utility (e.g. revenue, profit, ROI, …) instead. That’s the definition of the rational decision, right?
Hmm, for some reason you now seem to have several more problems.
#DecisionTheory #Optimization #rationality #AppliedML -
Me - Kobi, have you decided who you are supporting in the final?
Kobi - Paris St Germain
Me (roll eyes)
Kobi - I mean England. I mean Argentina. Or France.
-
We know inflation impacts expected (monetary) value, but how might it impact expected (subjective) utility?
Are there standard discount rates for subjective utility (over time) that are analogous to inflation rates on expected (monetary) value?
Are there illustrative real world examples?
Have the subjective discount rates been formalized somewhere?
I considered time discounting (e.g., hyperbolic discounting), but wasn't sure if that's sufficiently analogous.
-
We know inflation impacts expected (monetary) value, but how might it impact expected (subjective) utility?
Are there standard discount rates for subjective utility (over time) that are analogous to inflation rates on expected (monetary) value?
Are there illustrative real world examples?
Have the subjective discount rates been formalized somewhere?
I considered time discounting (e.g., hyperbolic discounting), but wasn't sure if that's sufficiently analogous.
-
Have been thinking for some time about whether it's permissible to bring about a #TransformativeExperience for someone else when that person doesn't know or resists it happening.
I mean transformative experience in the technical philosophical sense from LA Paul: and experience by our values and preferences change. On Paul's account, it already looks like we can't make a #RationalDecision about whether to choose a trans. exp. (on regular #DecisionTheory), but what about on behalf of others?
-
"Functional Decision Theory: A New Theory of Instrumental Rationality" by Yudkowsky and Soares
:arxiv: https://arxiv.org/abs/1710.05060
A fascinating paper introducing a new decision theory which solves many problems found on Causal Decision Theory and Evidential Decision Theory.
The idea behind FDT is to fix one's decision process as the output of a fixed mathematical function which answer the question "Which output of this very function would yield the best outcome?"
-
"Functional Decision Theory: A New Theory of Instrumental Rationality" by Yudkowsky and Soares
:arxiv: https://arxiv.org/abs/1710.05060
A fascinating paper introducing a new decision theory which solves many problems found on Causal Decision Theory and Evidential Decision Theory.
The idea behind FDT is to fix one's decision process as the output of a fixed mathematical function which answer the question "Which output of this very function would yield the best outcome?"
-
Recently came across this argument and love it, click the link if you want to know why you should choose worse options.
-
Recently came across this argument and love it, click the link if you want to know why you should choose worse options.
-
I use apps like this for academic community, esp. #academicphilosophy.
If you are a student, teacherer, research, faculty around #academic #philosophy of any kind and/or #gender (i.e. gender and women studies, womens' studies, gender studies, etc.), #sts, #legaltheory and #jurisprudence, #criticalrace, #historyofscience, #politicaltheory, #decisiontheory, #disabilitystudies, or something akin to these areas in the social sciences + humanities, I want to find you and get your toots in my feed!
-
Ahhh I'm doing it, I'm finally doing it! Here's my #introduction + first toot:
I'm an Assistant Professor and health economist at #UMass-Amherst. I apply #decisiontheory to #healthcare -- basically, I try to understand how physicians make decisions for patients and how these decisions can be improved.
I like canines, coffee, cooking, and cocktails. Also everything fantasy-related (books/movies/ video games).
-
OK, time for a proper #introduction
I am an #economist, with interests in (mostly dynamic) #gametheory and #decisiontheory. But I dabble in other things as well.
I am very interested in the interface between Economics and #CS, though I have yet to produce meaningful work there...
I use #tex for my writing; #lyx too sometimes. A few years ago I wrote the LaTeXtools plugin for the Sublime Text editor, with help from many contributors of course. Others are now carrying the torch!
-
OK, time for a proper #introduction
I am an #economist, with interests in (mostly dynamic) #gametheory and #decisiontheory. But I dabble in other things as well.
I am very interested in the interface between Economics and #CS, though I have yet to produce meaningful work there...
I use #tex for my writing; #lyx too sometimes. A few years ago I wrote the LaTeXtools plugin for the Sublime Text editor, with help from many contributors of course. Others are now carrying the torch!
-
Noticed the name of Rapoport in @neauoire 's wiki and because he was mentioned on #decisionTheory class, so I quickly looked up what he did, and gosh, why the heck did our teacher not talk more about his work???