Thank you for posting this! I’m posting here for the first time, although I’ve spent a significant amount of time reading the Sequences already (I just finished Seeing with Fresh Eyes). The comments on determinism cleared up a few uncertainties about Newcomb’s Problem for me.
When I have explained the problem to others, I have usually used the phrasing where Alpha is significantly better than average at predicting what you will choose, but not perfect. (This helps reduce incredulity on the part of the average listener.) I have also used the assumption that Alpha does this by examining your mental state, rather than by drawing causal arrows backward in time. One of my friends suggested precommitting to a strategy that one-boxes 51% of the time and two-boxes 49% of the time, chosen at the time you receive the boxes by some source that is agreed to be random such as rolling two d10′s. His logic is that Alpha would probably read your mind accurately, and that if he did, he would decide based on your mental state to put the money in the box, since you are more likely to one-box than not.
This seemed like a very good strategy (assuming the logic and the model of the problem are correct, which is far from certain), and I wondered why this strategy wasn’t at least being discussed more. It seems that most other people were assuming determinism while I was assuming libertarian free will.
What do all of you think of my friend’s strategy?
Is the assumption of determinism a comment on the actual state of the universe, or simply a necessary assumption to make the problem interesting?
Well that would work for a predictor that 100% predicts your most likely strategy. If the predictor has even a slight chance of predicting the 49% strategy instead of the 51% strategy, you’ll lose out as you’re risking a million to gain a thousand. But yes, the discussion in my post assumes that the predictor can predict any sources of randomness that you have access to.
Thank you for posting this! I’m posting here for the first time, although I’ve spent a significant amount of time reading the Sequences already (I just finished Seeing with Fresh Eyes). The comments on determinism cleared up a few uncertainties about Newcomb’s Problem for me.
When I have explained the problem to others, I have usually used the phrasing where Alpha is significantly better than average at predicting what you will choose, but not perfect. (This helps reduce incredulity on the part of the average listener.) I have also used the assumption that Alpha does this by examining your mental state, rather than by drawing causal arrows backward in time. One of my friends suggested precommitting to a strategy that one-boxes 51% of the time and two-boxes 49% of the time, chosen at the time you receive the boxes by some source that is agreed to be random such as rolling two d10′s. His logic is that Alpha would probably read your mind accurately, and that if he did, he would decide based on your mental state to put the money in the box, since you are more likely to one-box than not.
This seemed like a very good strategy (assuming the logic and the model of the problem are correct, which is far from certain), and I wondered why this strategy wasn’t at least being discussed more. It seems that most other people were assuming determinism while I was assuming libertarian free will.
What do all of you think of my friend’s strategy?
Is the assumption of determinism a comment on the actual state of the universe, or simply a necessary assumption to make the problem interesting?
Well that would work for a predictor that 100% predicts your most likely strategy. If the predictor has even a slight chance of predicting the 49% strategy instead of the 51% strategy, you’ll lose out as you’re risking a million to gain a thousand. But yes, the discussion in my post assumes that the predictor can predict any sources of randomness that you have access to.