Hi all, Sorry it’s been a while since I’ve been posting regularly. I’m officially committing to writing 1 post per week. So, give me a hard time if I forget about this.
One thing I’ve been thinking about a lot lately is objective functions (i.e., “goals”). In online 2-player games in computer science, we have to make a choice about how to model the other player.
- One classic choice is an adaptive adversary: another player that sees your moves, and then acts to try to make you lose.
- Another classic choice is an oblivious adversary: an adversary that doesn’t look at your moves, they fix their moves in advance.
- One less common choice is an ally. Someone who is happy when things are going well for you, and possibly even willing to endeavor to steer reality towards configurations where you have larger utility.
Now, there is a time and place for all of these models I’m sure. But I want to start considering the ally model more.
When your prior has large mass on the “other ppl are my allies” branch, it can cause the following things:
- generally leads to more honesty / open communication with others
- probably just leads to nicer interactions over all
- also hopefully it helps me remember that I have Terms for others in my utility function.
Now please, don’t take this too far. The universe does not care about you at all. Sufficiently large faceless organizations maybe also don’t super care about you, or at least have to overcome a really large amount of inertia to care about you or something. Really, please don’t interpret this post as advocating you to do something other than seek the truth. But I’m noticing that a personal failing of mine is to underestimate how much other people are willing / happy to help me out.
Q: ok Alek, that’s really vague. Do you have an example?
Sure fine. EX: \
One time (actually more like 3000 times) I was feeling bad and not fun to interact with. But then a friend talked to me anyways, even if they had to partially carry the conversation.
(oh yeah, thanks for that btw.)
Anyways, maybe above example is just evidence for the central claim of CBT (when you’re feeling down, you tend to distort reality and underestimate the potential goodness of everything).
So I guess my message for today is, if you have reason to suspect that you systematically negatively distort reality, then be a freaking efficient market and update your beliefs!