Diabloto96 | v1.6.0Mar 19th 2023 | |||
plex | v1.5.0Sep 16th 2021 | added link to FDT post | ||
Ruby | v1.4.0Oct 1st 2020 | (+2936/-241) | ||
brook | v1.3.0Aug 11th 2020 | (+66/-11) | ||
Ruby | v1.2.0Apr 11th 2020 | (+143/-24) | ||
Ruby | v1.1.0Apr 11th 2020 | (+253/-2922) | ||
Caspar Oesterheld | v0.0.44Oct 6th 2017 | (+80) | ||
Deku-shrub | v0.0.43May 12th 2017 | (+26/-50) /* Commonly discussed decision theories */ | ||
[anonymous] | v0.0.42May 12th 2017 | /* Commonly discussed decision theories */ | ||
[anonymous] | v0.0.41May 12th 2017 | (-30) /* Commonly discussed decision theories */ |
Decision theory is the study of principles and algorithms for making correct decisions—decisions—that is, decisions that allow an agent to achieve better outcomes with respect to its goals [1].
See also: Game Theory, Robust Agents, Utility Functions
The best known decision theories are Causal Decision Theory (CDT) and Evidential Decision Theory (EDT). On LessWrong, an alternative family of decision theories has been discussed heavily with the varying names of Updateless Decision Theory (UDT), Timeless Decision Theory (TDT), and Functional Decision Theory (FDT).
Decision theory is the formal study of principlesan agent's choices [1].
The best known decision theories are Causal Decision Theory (CDT) and algorithms for making correct decisions—that is, decisions that allowEvidential Decision Theory (EDT). On LessWrong, an agent to achieve better outcomes with respect to its goals. Every action at least implicitly represents a decision under uncertainty: in a state of partial knowledge, something has to be done, even if that something turns out to be nothing (call it "the null action"). Even if you don't know how you make decisions, decisions do get made, and so there has to be some underlying mechanism. What is it? And how can it be done better? Decision theory has the answers.
A core idea in decision theory is that of expected utility maximization, usually intractable to directly calculate in practice, but an invaluable theoretical concept. An agent assigns utility to every possible outcome: a real number representing the goodness or desirability of that outcome. The mapping of outcomes to utilities is called the agent's utility function. (The utility function is said to be invariant under affine transformations: that is, the utilities can be scaled or translated by a constant while resulting in all the same decisions.) For every action that the agent could take, sum over the utilities of the various possible outcomes weighted by their probability: this is the expected utility of the action, and the action with the highest expected utility is to be chosen.
The limitations and pathologiesalternative family of decision theories can be analyzed by consideringhas been discussed heavily with the decisions they suggest in the certain idealized situations that stretch the limitsvarying names of decision theory's applicability. Some of the thought experiments more frequently discussed on LW include:
Standard theories well-known in academia:
Theories invented by researchers associated with MIRI and LW:
Other decision theories are listed in A comprehensive list of decision theories (FDT).
Other decision theories are listed in A comprehensive list of decision theories.
Decision theory is the study of principles and algorithms for making correct
decisions—decisions—that is, decisions that allow an agent to achieve better outcomes with respect to itsgoals [1]goals. Every action at least implicitly represents a decision under uncertainty: in a state of partial knowledge, something has to be done, even if that something turns out to be nothing (call it "the null action"). Even if you don't know how you make decisions, decisions do get made, and so there has to be some underlying mechanism. What is it? And how can it be done better? Decision theory has the answers.See also:Note: this page needs to be updated with content regarding Functional Decision Theory, the latest theory from MIRI.Related: Game Theory, Robust Agents, Utility Functions
A core idea in decision theory is that of expected utility maximization, usually intractable to directly calculate in practice, but an invaluable theoretical concept. An agent assigns utility to every possible outcome: a real number representing the goodness or desirability of that outcome. The mapping of outcomes to utilities is called the agent's utility function. (The utility function is said to be invariant under affine transformations: that is, the utilities can be scaled or translated by a constant while resulting in all the same decisions.) For every action that the agent could take, sum over the utilities of the various possible outcomes weighted by their probability: this is the expected utility of the action, and the action with the highest expected utility is to be chosen.
Thought experiments
The
bestlimitations and pathologies of decision theories can be analyzed by considering the decisions they suggest in the certain idealized situations that stretch the limits of decision theory's applicability. Some of the thought experiments more frequently discussed on LW include:Commonly discussed decision theories
Standard theories well-known in academia:
Theories invented by researchers associated with MIRI and LW:
Other decision theories are listed in A comprehensive list of decision theories.
Blog posts
Sequence by AnnaSalamon
Sequence by orthonormal (Decision Theories: A Semi-Formal Analysis)
(CDT)and SubstitutionSee also
(EDT). On LessWrong, an alternative family oftheories has been discussed heavily with the varying names oftheory, UpdatelessDecision Theory (UDT), Timeless Decision Theory (TDT), and Functional Decision Theory (FDT).decision theory