= 27d567bc2d48d9f40e9ccc20ea370b15)
handoflixue comments on Rationality, Transhumanism, and Mental Health - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (129)
You would probably benefit from learning to evaluate "the utility of this situation, SANS the risk*utility of losing face".
i.e. "By giving an honest reply, I might lose face (UNEVALUATED/NULL utility) I might also gain karma (positive utility), and I might get useful suggestions (positive utility). Okay, now that we've established this is positive utility, we're going to go do it."
The key part is having your "algorithm" identify that "Calculate RiskUtility of Losing Face" is a faulty module, and stop calling that algorithm. I *personally find it useful to think "Okay, I have real trouble with the math for this utility, so I'll save it for last - if I'm still on the fence, it can be the tie breaker, but if everything else clearly points towards X outcome, I will just do X and not worry about it."
That's actually a BIG part of what's broken - it's got it's own event-handler.
I'm actually trying to point out a more generic problem than my own personal woes, with my original post:
Sometimes, you have processes running that simply corrupt your utility tables or your probability tables, and you don't have a good strategy to correct.
In Game Theory, one of the things that can be modeled is "signaling errors" - errors that, when they occur, cause you to perform a move that is different than the move you would have chosen.
What happens when your learning algorithm has "transcription errors" - errors that, when they occur, cause your utility tables or your probability tables to dangerously corrupt?
Worse, how do you construct a strategy to correct this process, when your "construct a strategy" process is throwing errors?
Here's a suggestion, and I'm not sure whether it's safe for you: hold off on developing strategies. Give yourself time to observe what you're doing. There's interesting and important stuff in the moments you're skipping over.
For me, it was really important to learn to ask, "What am I doing?" This is a very neutral research question.
It is NOT "What am I doing wrong? It is NOT "What can I do right now to fix things?"
For you, a specific question could be "What am I doing in the moment when I choose strategies?" If that's hard to focus on, choose something easier.
Psychological work is like doing original research.
That's somewhat hard to answer, anymore. I used to spend inordinate amount of time doing self-analysis, especially in emotionally intense situations, but any more, whenever I try to examine myself, this brain-fog rolls in and I can't think clearly. But it clearly seems the important question, so I don't know what "easier" thing I could focus on, that would be at all relevant.
I'm beginning to wonder if you've got a case of Robot's Revenge (a notion which I think is from RA Wilson, but I can't find a cite). Robot's Revenge is what can happen when you try to make yourself do something you just don't want to do, and you find yourself forgetting, making mistakes, and generally unable to be effective at obeying. Wilson's description implied that it was something which could happen when someone else is giving the orders, but I don't see why it couldn't happen internally.
This fits with a general model I've been developing that you've done a lot of being harsh with yourself, and aren't letting natural systems that help you function do their work.
Important note: this is all hypothetical. I'm not a therapist and I don't have studies backing me up, either. All I've got is a lot of self-observation, and some recent improvement in my levels of depression and inertia.
This doesn't stop me from giving advice, of course. And my current notion is that you'd benefit from giving up on self-improvement for at least a month. Possibly for a year.
I've tried to do exactly that, actually. The result is no employment, no medical insurance, and no social safety net. Our culture is rapidly losing patience with my inability to man up and earn my keep.
What I had in mind-- and it may be unfeasible-- was to do what you can for your situation without trying to revise yourself.
I'm suspecting that you and I mean different things when we say the same words-- not surprising when we're talking about somewhat unusual psychological work.. What happened when you gave up on self-improvement?
Yes, we definitely mean different things when we say the same words - I've always embraced an 'extended' view of self, where my self and my situation are utterly inseparable. In my ontology, the phrase "do what you can for your situation without trying to revise yourself" literally has no meaning.
in a word, entropy.
To elaborate, I stopped trying to tweak my psychology so that getting up in the morning could be bearable, stopped expending willpower and brainpower to seek delayed gratification, stopped playing mental contortions to emulate hope, and stopped putting forth a mask to everyone around me that everything was okay. In response, what meager social networks I had established eroded, what meager opportunities I had to feed and house myself eroded, and what meager opportunities I had to quell the screaming in my head eroded, and I eventually settled into a new, lower-energy ground state.