What’s your take on playing a PD against someone who is implementing a different decision algorithm to the one you are implementing, albeit strongly (logically) correlated in terms of outputs?
I’m guessing that “strongly logically correlated in terms of outputs” means that it has the same outputs for a large fraction of inputs (but not all) according to some measure over the space of all possible inputs.
If that’s all you know, then there will likely be nearly zero logical correlation between your outputs for this instance, and what you will decide depends upon what your decision algorithm does when there is close to zero logical correlation.
If you have more specific information than just existence of a strong logical correlation in general, then you should use it. For example, you may be told that the measure over which the correlation is taken is heavily weighted toward your specific inputs for this instance, and that the other player is given the same inputs. That raises the logical correlation between outputs for this instance, and (if your decision algorithm depends upon such things) you should cooperate.
I had something like the following in mind: you are playing the PD against someone implementing “AlienDT” which you know nothing about except that (i) it’s a completely different algorithm to the one you are implementing, and (ii) that it nonetheless outputs the same action/policy as the algorithm you are implementing with some high probability (say 0.9), in a given decision problem.
It seems to me that you should definitely cooperate in this case, but I have no idea how logi-causalist decision theories are supposed to arrive at that conclusion (if at all).
This is why I suggested naming FDT “functional decision theory” rather than “algorithmic decision theory”, when MIRI was discussing names.
Suppose that Alice is an LDT Agent and Bob is an Alien Agent. The two swap source code. If Alice can verify that Bob (on the input “Alice’s source code”) behaves the same as Alice in the PD, then Alice will cooperate. This is because Alice sees that the two possibilities are (C,C) and (D,D), and the former has higher utility.
The same holds if Alice is confident in Bob’s relevant conditional behavior for some other reason, but can’t literally view Bob’s source code. Alice evaluates counterfactuals based on “how would Bob behave if I do X? what about if I do Y?”, since those are the differences that can affect utility; knowing the details of Bob’s algorithm doesn’t matter if those details are screened off by Bob’s functional behavior.
The same holds if Alice is confident in Bob’s relevant conditional behavior for some other reason, but can’t literally view Bob’s source code. Alice evaluates counterfactuals based on “how would Bob behave if I do X? what about if I do Y?”, since those are the differences that can affect utility; knowing the details of Bob’s algorithm doesn’t matter if those details are screened off by Bob’s functional behavior.
Hm. What kind of dependence is involved here? Doesn’t seem like a case of subjunctive dependence as defined in the FDT papers; the two algorithms are not related in any way beyond that they happen to be correlated.
Alice evaluates counterfactuals based on “how would Bob behave if I do X? what about if I do Y?”, since those are the differences that can affect utility...
Sure, but so do all agents that subscribe to standard decision theories. The whole DT debate is about what that means.
What’s your take on playing a PD against someone who is implementing a different decision algorithm to the one you are implementing, albeit strongly (logically) correlated in terms of outputs?
I’m guessing that “strongly logically correlated in terms of outputs” means that it has the same outputs for a large fraction of inputs (but not all) according to some measure over the space of all possible inputs.
If that’s all you know, then there will likely be nearly zero logical correlation between your outputs for this instance, and what you will decide depends upon what your decision algorithm does when there is close to zero logical correlation.
If you have more specific information than just existence of a strong logical correlation in general, then you should use it. For example, you may be told that the measure over which the correlation is taken is heavily weighted toward your specific inputs for this instance, and that the other player is given the same inputs. That raises the logical correlation between outputs for this instance, and (if your decision algorithm depends upon such things) you should cooperate.
Depends on the decision algorithm! Do you have a specific one in mind?
E.g., LDT will defect against CDT.
I had something like the following in mind: you are playing the PD against someone implementing “AlienDT” which you know nothing about except that (i) it’s a completely different algorithm to the one you are implementing, and (ii) that it nonetheless outputs the same action/policy as the algorithm you are implementing with some high probability (say 0.9), in a given decision problem.
It seems to me that you should definitely cooperate in this case, but I have no idea how logi-causalist decision theories are supposed to arrive at that conclusion (if at all).
This is why I suggested naming FDT “functional decision theory” rather than “algorithmic decision theory”, when MIRI was discussing names.
Suppose that Alice is an LDT Agent and Bob is an Alien Agent. The two swap source code. If Alice can verify that Bob (on the input “Alice’s source code”) behaves the same as Alice in the PD, then Alice will cooperate. This is because Alice sees that the two possibilities are (C,C) and (D,D), and the former has higher utility.
The same holds if Alice is confident in Bob’s relevant conditional behavior for some other reason, but can’t literally view Bob’s source code. Alice evaluates counterfactuals based on “how would Bob behave if I do X? what about if I do Y?”, since those are the differences that can affect utility; knowing the details of Bob’s algorithm doesn’t matter if those details are screened off by Bob’s functional behavior.
Hm. What kind of dependence is involved here? Doesn’t seem like a case of subjunctive dependence as defined in the FDT papers; the two algorithms are not related in any way beyond that they happen to be correlated.
Sure, but so do all agents that subscribe to standard decision theories. The whole DT debate is about what that means.