Opinions expressed are my own and not endorsed by anyone.
Formerly @ ARC Evals aka METR
Opinions expressed are my own and not endorsed by anyone.
Formerly @ ARC Evals aka METR
I wonder how many recent trans people tried/considered doubling down on their sex (eg males taking more testosterone) instead first. Maybe (for some people) either end of gender spectrum is comfortable and being in the middle feels bad¿ Anybody know? Don’t want to ask my friends because this Q will certainly anger them
All games. Find where position is stored etc automatically i mean. It will certainly have failure cases. Easy to make a game that breaks it. The question is if an adversarial agent can easily break it in a regular (ie not adversarially chosen) game.
You may be right. Perhaps the way to view this idea is “yet another fuzzy-boundary RL helper technique” that works in a very different way and so will have different strengths and weaknesses than stuff like RLHF. So if one is doing the “serially apply all cheap tricks that somewhat reduce risk” approach then this can be yet another thing in your chain.
This is what I get for glossing over the math. RIP
Have you ever tried hiring someone or getting a job? Mostly lemons all around (apologies for the offense, jobseekers, i’m sure you’re not the lemon)
What about mediocre optimizers? Are they not worth fooling with?
What makes you say so? Seems like 25% chance possible to me. You can find where position is stored and watch for sudden changes. Same thing with score & inventory...
I do not propose one applies this method to a prepotence
I suspect that past therapists existed in your community and knew what you’re actually like so were better able to give you actual true information instead of having to digest only your bullshit and search for truth nuggets in it.
Furthermore, I suspect they didn’t lose their bread when they solve your problem! We have a major incentive issue in the current arrangement!
Andor is a word now. You’re welcome everybody. Celebrate with champagne andor ice cream.
Is it rude to make a new tag without also tagging a handful of posts for it? A few tags I kinda want:
explanation: thing explained.
idea: an idea for a thing someone could do (weaker version of “Research Agenda” tag)
stating the obvious: pointing out something obviously true but maybe frequently overlooked
experimental result
theoretical result
novel maybe: attempts to do something new (in the sense of novelty requirements for conference publications)
I would watch a ten hour video of this. (It may also be more persuasive to skeptics.)
I think I was wrong and you & Adele Lopez are right and pilot wave would be more lines. I am concerned about god’s RAM though… Maybe if they’ve got good hardware for low-rank matrices then it’s fine.
Furthermore, since pilot wave theory has no collapse postulate, it does not even get rid of the existence of multiple words.
Seems cruxy to me. I thought pilot wave did uhh somehow explicitly set those terms to zero, but I never actually saw that written down. But you can actually use the corpuscles to say which parts of the wave get zeroed out right? If two particles are actually entangled then their corpuscle equations or whatever will tell you that you can’t delete that part of the wave? (Apologies if these questions are malformed — I don’t actually know the math anymore — thought I wouldn’t need it to understand the interpretations but now I’m thinking maybe I do. Really, I need code, not math...)
What does the Elitzur–Vaidman bomb tester look like in Pilot Wave?
Could I just say that the wave interferes with the bomb (if it’s live) and bumps the particle after the wave hits the mirror?
(I don’t actually know enough physics to, like, do the math on that.)
Oh sweet
yud is always like “it cant be that if it werent so then you wouldnt have thought it werent, so it must not be not false”. Oh no my sub-yud IQ might be showing
I must abstain from further culturewaring but this thought experiment is blowing my mind. I hadn’t heard of it.
Oh I have 0% success with any long conversations with an LLM about anything. I usually stick to one question and rephrase and reroll a number of times. I am no pro but I do get good utility out of LLMs for nebulous technical questions