Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Comment author: cousin_it 13 June 2017 05:23:02PM *  4 points [-]

Very impressive, I'm happy that Paul ended up there! There's still a lot of neural network black magic though. Stuff like this:

We use standard settings for the hyperparameters: an entropy bonus of β = 0.01, learning rate of 0.0007 decayed linearly to reach zero after 80 million timesteps (although runs were actually trained for only 50 million timesteps), n = 5 steps per update, N = 16 parallel workers, discount rate γ = 0.99, and policy gradient using Adam with α = 0.99 and ε = 10−5.

For the reward predictor, we use 84x84 images as inputs (the same as the inputs to the policy), and stack 4 frames for a total 84x84x4 input tensor. This input is fed through 4 convolutional layers of size 7x7, 5x5, 3x3, and 3x3 with strides 3, 2, 1, 1, each having 16 filters, with leaky ReLU nonlinearities (α = 0.01). This is followed by a fully connected layer of size 64 and then a scalar output. All convolutional layers use batch norm and dropout with α = 0.5 to prevent predictor overfitting.

I know I sound like a retrograde, but how much of that is necessary and how much can be figured out from first principles?

Comment author: lahwran 14 June 2017 08:07:54AM 2 points [-]

seems like neural network anxiety misses the point of this paper - that the artificial intelligence algorithms that actually work can in fact be brought towards directions that have a shot at making them safe

Comment author: lahwran 13 May 2017 05:43:42PM 0 points [-]


Comment author: lahwran 22 April 2017 03:24:50PM 0 points [-]

this seems a bit oversold, but basically represents what I think is actually possible

Comment author: lahwran 09 April 2017 01:11:25PM 1 point [-]

I will say the same thing here that I did there: if (and only if) you attempt to kill me, I'll attempt to kill you back with appropriately much torture to make you fear that outcome. your morality should be selected for being the best for you, logically. I commit to making sure anything that involves attacking me is very bad for you.

Comment author: username2 09 April 2017 05:03:53AM 2 points [-]

Um, obvious solution: redefine your morality. There is no objective morality. If you think the net utility of world is negative, that really says more about you than the world.

And if you are totally sincere in this belief, then honestly: seek professional mental help.

Comment author: lahwran 09 April 2017 01:07:45PM 1 point [-]

for what it's worth, I don't think professional mental health is any good most of the time, and it's only worth it if you're actually psychotic. for things that don't totally destroy your agency and just mostly dampen it, I think just doing things on your own is better.

Comment author: Good_Burning_Plastic 05 April 2017 11:05:12PM 0 points [-]

I think people who say[1] that guess culture only exists some places are meaningfully confused.

Or maybe they just don't fall prey to the fallacy of gray and realize it sometimes might make sense to call something black even though it doesn't literally scatter exactly no light at all (otherwise there'd be no point in having a word if it didn't apply to anything at all).

Comment author: lahwran 06 April 2017 05:45:59AM 0 points [-]

I understand that. I wrote the post you're replying to with that in mind. I think the thing that people call guess culture actually applies almost everywhere, and anything but high trust between very close friends will secretly be only using different words, but have the same guessing patterns. I'm not making some wordplay claim here, I actually think there is a high magnitude error in the theory and that the update is to apply guess culture almost everywhere.

Comment author: RedMan 01 April 2017 11:24:06AM *  0 points [-]

I read this as assuming that all copies deterministically demonstrate absolute allegiance to the collective self. I question that assertion, but have no clear way of proving the argument one way or another. If 're-merging' is possible, mergeable copies intending to merge should probably be treated as a unitary entity rather than individuals for the sake of this discussion.

Ultimately, I read your position as stating that suicide is a human right, but that secure deletion of an individual is not acceptable to prevent ultimate harm to that individual, but is acceptable to prevent harm caused by that individual to others.

This is far from a settled issue, and has analogy in the question 'should you terminate an uncomplicated preganancy with terminal birth defects?' Anencephaly is a good example of this situation. The argument presented in the OP is consistent with a 'yes', and I read your line of argument as consistent with a clear 'no'.

Thanks again for the food for thought.

Comment author: lahwran 05 April 2017 07:28:55PM 0 points [-]

I acausally cooperate with agents who I evaluate to be similar to me. That includes most humans, but it includes myself REALLY HARD, and doesn't include an unborn baby. (because babies are just templates, and the thing that makes them like me is being in the world for a year ish.)

Comment author: Good_Burning_Plastic 02 April 2017 09:11:56PM 1 point [-]

I don't think you can escape guess culture.

Sometimes you can escape it literally, e.g. move to a different city or find a different social circle.

Comment author: lahwran 05 April 2017 07:25:39PM *  3 points [-]

No, I mean this more strongly than that. I literally do not think it's possible to interact with humans without using guess culture. all of human interaction is the same as the thing that got labeled guess culture; sometimes it can also be ask culture or also tell culture, but I think it's meaningfully true that ask culture is just different defaults for guess culture, and you're still actually doing just as much guessing. I do think you can reduce guessing via trust, and that guessing with the goal of building trust is something people automatically do, but I think people who say[1] that guess culture only exists some places are meaningfully confused.

I also think that "tell culture" is a subset of interaction that only works with high trust - and I think people who "are" guess culture will naturally do things that look more like tell culture when trust is high.

[1] (or have said in the past, before updating on my saying this)

Comment author: blacktrance 01 April 2017 09:42:07PM *  0 points [-]

Upon further consideration, it seems to me that while it being enforced can make it worse, much of the prosociality cluster (e.g. guess culture) is oppressive in itself.

Comment author: lahwran 02 April 2017 06:47:50AM 3 points [-]

I don't think you can escape guess culture. you only can pretend you don't have it, and then pay the price.

Comment author: RedMan 29 March 2017 12:05:21AM 0 points [-]

Correct, that is different from the initial question, you made your position on that topic clear.

Would the copy on the satellite disagree about the primacy of the copy not in the torture sim? Would a copt have the right to disagree? Is it morally wrong for me to spin up a dozen copies of myself and force them to fight to the death for my amusement?

I'm guessing based on your responses that you would agree with the statement 'copies of the same root individual are property of the copy with the oldest timestamped date of creation, and may be created, destroyed, and abused at the whims of that first copy, and no one else'

If you copy yourself, and that copy commits a crime, are all copies held responsible, just the 'root' copy, or just the 'leaf' copy?

Thank you for the challenging responses!

Comment author: lahwran 29 March 2017 07:27:48AM 1 point [-]

no. copies are all equally me until they diverge greatly; I wouldn't mind 10 copies existing for 10 minutes and then being deleted any more than I would mind forgetting an hour. the "primary copy" is maybe a bad way to put it; I only meant that colloquially, in the sense that looking at that world from the outside, the structure is obvious.

  1. copy on the satellite would not disagree
  2. yes would have the right, but as an FDT agent a copy would not disagree except for straight up noise in the implementation of me; I might make a mistake if I can't propagate information between all parts of myself but that's different
  3. that sounds kind of disgusting to experience as the remaining agent, but I don't see an obvious reason it should be a moral thing. if you're the kind of agent that would do that, I might avoid you
  4. copies are not property, they're equal
  5. that's very complicated based on what the crime is and the intent of the punishment/retribution/restorative justice/etc

View more: Next