Gram_Stone

Karma: 1,785

Gram_Stone 18 Sep 2017 14:03 UTC
5 points
on: LW 2.0 Strategic Overview
Will there be LaTeX support?

Gram_Stone 28 Jun 2017 22:09 UTC
1 point
on: What useless things did you understand recently?
(Not very familiar with math.)

The Heyting-algebraic definition of implication makes intuitive sense to me, or at least after you state your confusion. ‘One circle lies inside the other’ is like saying A is a subset of B, which is a statement that describes a relation between two sets, and not a statement that describes a set, so we shouldn’t expect that that mental image would correspond to a set. Furthermore, the definition of implication you’ve given is very similar to the material implication rule; that we may substitute ‘P implies Q’ with ‘not-P or Q’.

Also, I have personally been enjoying your recent posts with few prerequisites. (Seems to be a thing.)

Gram_Stone 9 Jun 2017 11:55 UTC
1 point
on: Humans are not agents: short vs long term
I have what feels like a naive question. Is there any reason that we can’t keep appealing to even higher-order preferences? I mean, when I find that I have these sorts of inconsistencies, I find myself making an additional moral judgment that tries to resolve the inconsistency. So couldn’t you show the human (or, if the AI is doing all this in its ‘head’, a suitably accurate simulation of the human) that their preference depends on the philosopher that we introduce them to? Or in other cases where, say, ordering matters, show them multiple orderings, or their simulations’ reactions to every possible ordering where feasible, and so on. Maybe this will elicit a new judgment that we would consider morally relevant. But this all relies on simulation, I don’t know if you can get the same effect without that capability, and this solution doesn’t seem even close to being fully general.

I imagine that this might not do much to resolve your confusion however. It doesn’t do much to resolve mine.

Gram_Stone 8 Jun 2017 22:35 UTC
3 points
on: Bring up Genius

Discipline, especially internal psychological, also increases skills.

This is a little ambiguous; does he mean self-control or punishment?

Gram_Stone 5 Jun 2017 14:24 UTC
3 points
on: Cognitive Science/Psychology As a Neglected Approach to AI Safety
I think these are all points that many people have considered privately or publicly in isolation, but that thus far no one has explicitly written them down and drawn a connection between them. In particular, lots of people have independently made the observation that ontological crises in AIs are apparently similar to existential angst in humans, ontology identification seems philosophically difficult, and so plausibly studying ontology identification in humans is a promising route to understanding ontology identification for arbitrary minds. So, thank you for writing this up; it seems like something that needed to be written quite badly.

Some other problems that might be easier to tackle from this perspective include mind crime, nonperson predicates, and suffering risk, especially subproblems like suffering in physics.

Gram_Stone 3 Jun 2017 22:31 UTC
6 points
on: Rationality Quotes April—June 2017

We had succeeded in obtaining John yon Neumann as keynote speaker. He discussed the need for, and likely impact of, electronic computing. He mentioned the “new programming method” for ENIAC and explained that its seemingly small vocabulary was in fact ample: that future computers, then in the design stage, would get along on a dozen instruction types, and this was known to be adequate for expressing all of mathematics. (Parenthetically, it is as true today as it was then that “programming” a problem means giving it a mathematical formulation. Source languages which use “plain English” or other appealing vocabularies are only mnemonic disguises for mathematics.) Von Neumann went on to say that one need not be surprised at this small number, since about 1,000 words were known to be adequate for most situations of real life, and mathematics was only a small part of life, and a very simple part at that. This caused some hilarity in the audience, which provoked von Neumann to say: “If people do not believe that mathematics is simple, it is only because they do not realize how complicated life is.”

Franz L. Alt, “Archaeology of computers: Reminiscences, 1945--1947”, Communications of the ACM, volume 15, issue 7, July 1972, special issue: Twenty-fifth anniversary of the Association for Computing Machinery, p. 694. PDF.

Gram_Stone 3 Jun 2017 18:56 UTC
1 point
in reply to: ieltscertificate77’s comment on: Book recommendation requests
spam

Gram_Stone 2 Jun 2017 14:26 UTC
2 points
in reply to: simbyotic’s comment on: Book recommendation requests

Neuroscience of art & art appreciation

A slightly broader keyword would be ‘neuroaesthetics.’

Evolutionary basis for storytelling

I haven’t done an exhaustive literature search, but one book I’m going through right now is Brian Boyd’s On the Origin of Stories: Evolution, Cognition, and Fiction.

Gram_Stone 1 Jun 2017 23:02 UTC
5 points
in reply to: ChristianKl’s comment on: Book recommendation requests
psychopathology* (Genuinely trying to be helpful, not nitpicky; keywords are important.)

Related, broader keyword: abnormal psychology.

Gram_Stone 14 May 2017 21:27 UTC
5 points
on: Gears in understanding
Tangentially, I thought you might find repair theory interesting, if not useful. Briefly, when students make mistakes while doing arithmetic, these mistakes are rarely the effect of a trembling hand; rather, most such mistakes can be explained via a small set of procedural skills that systematically produce incorrect answers.

Gram_Stone 27 Apr 2017 20:19 UTC
0 points
in reply to: hamnox’s comment on: Background Reading: The Real Hufflepuff Sequence Was The Posts We Made Along The Way
Do you think I disagree with that?

Gram_Stone 27 Apr 2017 1:15 UTC
1 point
on: Background Reading: The Real Hufflepuff Sequence Was The Posts We Made Along The Way
I’ve had a strong urge to ask about the relation between Project Hufflepuff and group epistemic rationality since you started writing this sequence. This also seems like a good time to ask because your criticism of the essays that you cite (with the caveat that you believe them to contain grains of truth) seems fundamentally to be an epistemological one. Your final remarks are altogether an uncontroversial epistemological prescription, “We have time and we should use it because other things equal taking more time increases the reliability of our reasoning.”

So, if I take it that your criticism of the lack of understanding in this area is an epistemological one, then I can imagine this sequence going one of two ways. The one way is that you’ll solve the problem, or some of it, with your individual epistemological abilities, or at least start on this and have others assist. The other way is that before discussing culture directly, you’ll discuss group epistemic rationality, bootstrapping the community’s ability to reason reliably about itself. But I don’t really like to press people on what’s coming later in their sequence. That’s what the sequence is for. Maybe I can ask some pointed questions instead.

Do you think group epistemic rationality is prior to the sort of group instrumental rationality that you’re focusing on right now? I’m not trying to stay hyperfocused on epistemic rationality per se. I’m saying that you’ve demonstrated that the group has not historically done well in an epistemological sense on understanding the open problems in this area of group instrumental rationality that you’re focusing on right now, and now I’m wondering if you, or anyone else, think that’s just a failure thus far that can be corrected by individual epistemological means only and distributed to the group, or if you think that it’s a systemic failure of the group to arrive at accurate collective judgments. Of course it’s hardly a sharp dichotomy. If one thought the latter, then one might conclude that it is important to recurse to social epistemology for entirely instrumental reasons.

If group epistemic rationality is not prior to the sort of instrumental rationality that you’re focusing on right now, then do you think it would be nevertheless more effective to address that problem first? Have you considered that in the past? Of course, it’s not entirely necessary that these topics be discussed consecutively, as opposed to simultaneously.

How common do you think knowledge of academic literature relevant to group epistemic rationality is in this group? Like, as a proxy, what proportion of people do you think know about shared information bias? The only sort of thing like this I’ve seen as common knowledge in this group is informational cascades. Just taking an opportunity to try and figure out how much private information I have, because if I have a lot, then that’s bad.

How does Project Hufflepuff relate to other group projects like LW 2.0/the New Rationality Organization, and all of the various calls for improving the quality of our social-epistemological activities? I now notice that all of those seem quite closely focused on discussion media.

Gram_Stone 23 Apr 2017 18:21 UTC
3 points
on: The AI Alignment Problem Has Already Been Solved(?) Once
I enjoyed this very much. One thing I really like is that your interpretation of the evolutionary origin of Type 2 processes and their relationship with Type 1 processes seems a lot more realistic to me than what I usually see. Usually the two are made to sound very adversarial, with Type 2 processes having some kind of executive control. I’ve always wondered how you could actually get this setup through incremental adaptations. It doesn’t seem like Azathoth’s signature. I wrote something relevant to this in correspondence:

If Type 2 just popped up in the process of human evolution, and magically got control over Type 1, what are the chances that it would amount to anything but a brain defect? You’d more likely be useless in the ancestral environment if a brand new mental hierarch had spontaneously mutated into existence and was in control of parts of a mind that had been adaptive on their own for so long. It makes way more sense to me to imagine that there was a mutant who could first do algorithmic cognition, and that there were certain cues that could trigger the use of this new system, and that provided the marginal advantage. Eventually, you could use that ability to make things safe enough to use the ability even more often. And then it would almost seem like it was the Type 2 that was in charge of the Type 1, but really Type 1 was just giving you more and more leeway as things got safer.

Gram_Stone 5 Apr 2017 11:40 UTC
12 points
on: Brief update on the consequences of my “Two arguments for not thinking about ethics” (2014) article
Thank you for following up after all this time. Longitudinal studies seem important.

Gram_Stone 17 Mar 2017 13:29 UTC
0 points
in reply to: Alicorn’s comment on: I Want To Live In A Baugruppe
This is neat.

Gram_Stone 9 Mar 2017 17:28 UTC
0 points
on: An Intuition on the Bayes-Structural Justification for Free Speech Norms
I find the metaphor plausible. Let’s see if I understand where you’re coming from.

I’ve been looking into predecision processes as a means of figuring out where human decisionmaking systematically goes wrong. One such process is hypothesis generation. I found an interesting result in this paper; the researchers compared the hypothesis sets generated by individuals, natural groups and synthetic groups. In this study, a synthetic group’s hypothesis set is agglomerated from the hypothesis sets of individuals who never interact socially. They found that natural groups generate more hypotheses than individuals, and that synthetic groups generate more hypotheses than either. It appears that social interaction somehow reduces the number of alternatives that a group considers relative to what the sum of their considerations would be if they were not a group.

Now, this could just be biased information search. One person poses a hypothesis aloud, and then the alternatives become less available to the entire group. But information search itself could be mediated by motivational factors, like if I write “one high-status person poses a hypothesis aloud...”, and this is now a hypothesis about biased information search and a zero-sum social-control component. It does seem worth noting that biased search is currently a sufficient explanation by itself, so we might prefer it by parsimony, but at this level of the world model, it seems like things are often multiply determined.

Importantly, creating synthetic groups doesn’t look like punishing memetic-warfare/social-control at all. It looks like preventing it altogether. This seems like an intervention that would be difficult to generate if you thought about the problem in the usual way.

Gram_Stone 7 Mar 2017 0:58 UTC
0 points
in reply to: Zack_M_Davis’s comment on: Am I Really an X?
I was familiar with this.

I find the first etiology similar to my model. Did you mean to imply this similarity by use of the word ‘indeed’? I can see how one might interpret my model as an algorithm that outputs a little ‘gender token’ black box that directly causes the self-reports, but I really didn’t mean to propose anything besides “Once gendered behavior has been determined, however that occurs, cisgender males don’t say “I’m a boy!” for cognitive reasons that are substantially different from the reasons that transgender males say “I’m a boy!” ” Writing things like “behaviorally-masculine girls” just sounds like paraphrase to me. Should it not? On the other hand, as I understand it, the second etiology substantially departs from this. In that case it is proposed that transgender people that transition later in life perform similar behaviors for entirely different cognitive reasons.

I’ll reiterate that I admit to the plausibility of other causes of self-report. I do find your confidence surprising, however. I realize the visible controversy is not much evidence that you’re wrong, because we would expect controversy either way. Do you have thoughts on Thoughts on The Blanchard/Bailey Distinction? I’d just like to read them if they exist.

Gram_Stone 5 Mar 2017 16:42 UTC
0 points
in reply to: bogus’s comment on: Am I Really an X?
This is an excellent summary of my argument! Thank you so much for compressing this into a soundbite!

Gram_Stone 5 Mar 2017 16:41 UTC
3 points
in reply to: Articulator’s comment on: Am I Really an X?

I think this topic is really only as political as you make it.

I did in fact decide not to reply to the grandparent because I estimated that it would cause less harm in this respect than replying. This article is intended to be a contribution to the philosophy of gender identity in the style of EY’s executable philosophy, and it is more directly a reply to lucidfox’s Gender Identity and Rationality. This topic was perfectly acceptable in 2010.
What links here?
- satt's comment on Am I Really an X? by Gram_Stone (8 Mar 2017 0:42 UTC; 2 points)

Am I Really an X?

Gram_Stone5 Mar 2017 0:11 UTC

14 points

766 comments4 min readLW link