adrusi

Karma: 131

adrusi 23 Dec 2017 18:19 UTC
1 point
in reply to: tempus’s comment on: Rationalist Politicians
If a nerd won the presidency, it wouldn’t be great because they would say “true” things. It would be great because they would actually be concerned with figuring out what is true. They might actually change their minds if they realized they were wrong.
If you agree with Trump, then let’s allow that he “says true things.” That doesn’t mean that he embodies what would be great about a nerd in the Oval Office. If Trump says true things, it’s because it gets him the support of certain segments of the population. If he had evidence that one of his beliefs was false, but his base still believed it, I’m quite certain he would go on professing the false belief. Now I don’t actually think Trump would even recognize his belief to be wrong in the presence of evidence, and I think a substantial fraction of politicians are little better than he.
The article didn’t claim that Obama was — rather, it claimed that Obama’s path to the presidency could be emulated by nerds.

adrusi 12 Jan 2018 16:23 UTC
5 points
on: Babble
I’ve been thinking about this same idea, and I thought your post captured the heart of the algorithm (and damn you for beating me to it 😉). But I think you got the algorithm slightly wrong, or simplified the idea a bit. The “babble” isn’t random, there are too many possible thoughts for random thought generation to ever arrive at something the prune filter would accept. Instead, the babble is the output of of a pattern matching process. That’s why computers have become good at producing babble: neural networks have become competent pattern matchers.
This means that the algorithm is essentially the hypothetical-deductive model from philisophy of science, most obvious when the thoughts you’re trying to come up with are explanations of phenomena: you produce an explanation by pattern matching, then prune the ones that make no goddamn sense (then if you’re doing science you take the explanations you can’t reject for making no sense and prune them again by experiment). That’s why I’ve been calling your babble-prune algorithm “psychological adbuctivism.”
Your babble’s pattern matching gets trained on what gets accepted by the prune filter, that’s why it gets better over time. But if your prune filter is so strict that it seldom accepts any of your babble’s output, your babble never improves. That’s why you must constrain the tyranny of your prune filter if you find yourself with nothing to say. If you never accept any of your babble, then you will never learn to babble better. You can learn to babble better by pattern matching off of what others say, but if your prune filter is so strict, you’re going to have a tough time finding other people who say things that pass your prune filter. You’ll think “thats a fine thing to say, but I would never say it, certainly not that way.” Moreover, listening to other people is how your prune filter is trained, so your prune filter will be getting better (that is to say, more strict) at the same time as your straggling babble generator is.
I’ve had success over the past year with making my prune filter less strict in conversational speech, and I think my babble has improved enough that I can put my prune filter back up to its original level. But I need to do the same with my writing and I find it harder to do that. With conversational speech, you have a time constraint, so if your prune filter is too strict you simply end up saying nothing — the other person will say something or leave before you come up with a sufficiently witty response. In writing, you can just take your time. If it takes you an hour toncome up with the next sentence, then you sit down and wait that god-forsaken hour out. You can get fine writing out with that process, but it’s slow. My writing is good enough that I never proofread (I could certainly still get something out of proofreading, but it isn’t compulsory, even for longer pieces of writing), but to get that degree of quality takes me forever, and I cant produce lower quality writing faster (which would be very useful for finishing my exams on time).

adrusi 12 Jan 2018 21:26 UTC
3 points
on: More Babble
This is a followup to my comment on the previous post.
This followup (Edit: alkjash’s followup post, not my followup comment) addresses my stated motivation for suggesting that the babble-generator is based on pattern-matching rather than a mere entropy. I had said that there are too many possible ideas for a entropy to generate reasonable ones. For babble to be produced by a random walk along the idea graph is more plausible. It’s not obvious that you couldn’t produce sufficiently high-quality babble with a random-walk along a well-constructed idea-graph.
Now, while I absolutely think the idea graph exists, and I agree that producing babble involves a walk along that graph, I am still committed to the belief that that walk is not random, but is guided by pattern matching. My first reason for holding this belief is introspection: I noticed that ideas are produced by “guess-and-check” (or babble-and-prune) by introspection, and I also noticed that the guessing process is based on pattern matching. That’s fairly weak evidence. My stronger reason for belieiving that babble is produced by pattern matching is that it’s safer to assume that a neurological process is based on pattern matching than random behavior. Neurons are naturally suited to forming pattern-matching machines (please forgive my lay-understanding of cognitive science), and while I don’t see why they couldn’t also form an entropy generator, I don’t suspect that a random walk down the idea graph would be more adaptive than a more “intelligent” pattern matching algorithm.
I also infer that the babble-generator is a pattern matcher from the predictions that makes. If the babble-generator is a random-walk down the idea-graph, then the only way to improve your babble should be to improve your idea graph. If the babble-generator is a pattern-matcher-diected-walk down the idea-graph then you should be able to improve your babble both by training the pattern-matcher well and by improving your idea-graph. Let’s say reading nonfiction impoves your idea graph more effectively than it trains a hypthetical pattern-matcher, and that writing novels trains your pattern-matcher more effectively than it improves your idea-graph. Then if the random-walk hypothesis is true, we should see the same kinds of improvements to babble when we read nonfiction and write novels, but if the pattern-matcher hypothesis is true we should expect different kinds of improvements.
***
I think for the most part we’re talking about the same thing, I’m just suggesting this additional detail of pattern-matching, which has normative consequences (as I sketched out in my previous comment). However I’m not quite sure that we’re talking about the same graphs. You say:
What is the Babble graph? It’s the graph within which your words and concepts are connected. Some of these connections are by rhyme and visual similarity, others are semantic or personal.
I certainly don’t think think that this graph is a graph of words, even though I agree that there can be connections representing syntactic relationships like rhyme. I don’t think that the babble algorithm is “start at some node of the graph, output the word associated with that node, then select a connected node and repeat.” There is an idea-graph, and it’s used in the production of babble, but not like that. I’m not sure if you were claiming that it does, but in case you were, I disagree. I would try to elaborate what role I do think the idea-graph plays in babble generation, but this comment is already getting very long.
I’m curious about the details of your model of this “babble-graph,” You mention that it can create new connections, which suggest to me that the “graph” is actually a static representation of an active process of connection-drawing. I could be convinced that the pattern-matching I’m talking about is actually a separate process which is responsible for forming these connections. But I’m fuzzy on what exactly you mean so I’m not sure that’s even coherent.
Great posts, I wouldn’t mind a part 3!

adrusi 12 Jan 2018 21:27 UTC
2 points
in reply to: alkjash’s comment on: Babble
I’ve made a reply to your followup.

adrusi 13 Jan 2018 0:24 UTC
1 point
in reply to: SquirrelInHell’s comment on: More Babble
Are you referring to the second half of my comment? Because perhaps I wasn’t clear enough. I’m confused what alkjash means, because some of their references to the babble graph seemed perfectly consistent with my understanding but I got the impression that overall we might not be talking about the same thing.if we are talking about the same thing then that whole section of my comment is irrelevant.

adrusi 13 Jan 2018 4:50 UTC
1 point
in reply to: alkjash’s comment on: More Babble
I’m inclined to think that the babble you’ve been describing is actually just thoughts, and not linguistic at all. You create thoughts by babble-and-prune and then a separate process converts the thoughts into words. I haven’t thought much about how that process works (and at first inspection I think it’s probably also structured as babble-and-prune), but I think it makes sense to think about it as separate.
If the processes of forming thoughts and phrasing them linguistically were happening at the same level, I’d expect it to be more intuitive to make syntax reflect semantics, like you see in Shakespeare where the phonetic qualities of a character’s speech reflect their personality. Instead, writing like that seems to require System 2 intervention.
But I must admit I’m biased. If I were designing a mind, I’d want to have thought generation uncoupled from sentence generation, but it doesn’t have to actually work that way.
Edit: If generating linguistic-babble happens on a separate level from generating thought-babble, then that has consequences for how to train thought-babble. Your suggestions of playing scrabble and writing haikus would train the wrong babble (nothing wrong with training linguistic-babble, that’s how you become a good writer, but I’m more interested in thought-babble). I think if you wanted to train thought-babble, you’d want to do something like freewriting or brainstorming — rapidly producing a set of related ideas without judgment.

One-Consciousness Universe

adrusi23 Jan 2018 8:46 UTC

0 points

18 comments2 min readLW link

(practicalontology.com)

adrusi 23 Jan 2018 15:56 UTC
5 points
in reply to: cousin_it’s comment on: One-Consciousness Universe
You’ve woven a story in which I am wrong, and it will be hard for me to admit that I am wrong. In doing so, you’ve made it tricky for to defend my point in the case that I’m not wrong.
You’re accusing my “conscion” of being the same kind of mysterious answer as phlogiston. It would be, if I were seriously proposing it as an answer to the mystery of consciousness. I’m not.
I view this one-electron universe model as an ontological koan. It makes us think “hey, reality could be this way rather than the way we think it is and we would be none the wiser — let’s try to deepen our understanding of reality in light of that.”
I’ll gladly concede a failure of my writing in not making it clearer that I’m not making any claim that the conscion exists, but rather that thinking about what it would mean for our understanding of consciousness if the conscion did exist, as described. I’m trying to force people to drop their intuition about the “flow” of consciousness. I’m saying that all our observations about consciousness can be equally well explained by this weird conscion hypothesis as can be by the conventional consciousness-as-the-christian-soul hypothesis, so we should notice that many of our intuitions about consciousness have simply been transplanted from theology, and we should not trust those.

adrusi 23 Jan 2018 16:25 UTC
−1 points
in reply to: cousin_it’s comment on: One-Consciousness Universe
I’m not trying to explain the theory of flow (not in this post, I do have some thoughts on the matter). I’m merely trying to induce doubt.
The conventional understanding of consciousness as the Christian soul doesn’t explain anything, really, just like the “conscion.” But because it’s tied up in millennia of Christian scholarship, there are suppositions attached to it that are indefensible.

adrusi 23 Jan 2018 17:23 UTC
1 point
in reply to: cousin_it’s comment on: One-Consciousness Universe
You posted this reply before I finished editing my previous comment to include its second clause, but I’ll respond as though the order were more natural.
That’s the wrong comparison to be making. Suppose the deist idea about the origin of the universe were dominant, and I proposed that God may not have created the universe. After all, deists, what created God? He was an unmoved mover? Well why couldn’t the universe have just been an unmoved movse in the first place? Sound like you’re just passing the recursive buck, deists! I’m not proposing any kind of better explanation, just offering a different non-explanation to induce doubt.
Doubt in what? Well I admit I don’t know all that much about deism, but let’s suppose that deists believed that even though god never intervened in the universe, he had intentions for how the universe should turn out, and it’s our job as his creations to honor his intentions like we would honor the intentions of our fathers. This baggage is not entailed by the core theory of deism, it just came along for the ride when deism evolved from older Christian metaphysics. That’s why even though my proposed alternative to deism is no more an explanation of the origin of the universe than deism is, it brings to attention the fact that that deism’s baggage is unnecessary and we should forsake it.
I’m not saying we need to doubt the conventional understanding of consciousness entirely, rather that we should recognize that it has baggage and forsake it. What’s the evidence that it has baggage? Well conventional intuition makes the idea of person-slices seem suspect, as I described in the post. Person-slices don’t seem suspect when you use the conscion model of consciousness. If the two hypotheses are equally non-explanatory, then it is baggage that causes the different intuition.

adrusi 23 Jan 2018 18:12 UTC
1 point
in reply to: Matthew Barnett’s comment on: One-Consciousness Universe
This is precisely the kind of gymnastics you need to do if you want to justify the foundational claim of altruism, that other people should matter to you. But what you’ve said is not sufficient to justify that. Why should I care about the person-slices the conscion visits if they are not my own?

adrusi 23 Jan 2018 21:35 UTC
3 points
in reply to: cousin_it’s comment on: One-Consciousness Universe
I’m not sure at this point what my goal was with this post, it would be too easy to fall into motivated reasoning after this back-and-forth. So I agree with you that my post fails to give evidence for “consciousness can be based on person-slices,” I just don’t know if I ever intended to give that positive conclusion.
I do think that person-slices are entirely plausible, and a very useful analytical tool, as Parfit found. I have other thoughts on consciousness which assume person-slices are a coherent concept. If this post is sufficient to make the burden of proof for the existence of person-slices not clearly fall to me, then it’s served a useful purpose.
***
By the way, I did give a positive account for the existence of person slices, comparing the notion of a person slice to something that we more readily accept exists:
What would it be like to be a person-slice? This seems to me to be analogous to asking “how can we observe a snapshot of an electron in time?” We can’t! Observation can only be done over an interval of time, but just because we can’t observe electron-slices doesn’t mean that we shouldn’t expect to be able to observe electrons over time, nor does the fact that we can observe electrons over time suggest that electron-slices are a nonsensical concept. Likewise, if there’s nothing it’s like to be a person-slice, that doesn’t mean that person-slices are nonsense.

adrusi 24 Jan 2018 15:08 UTC
5 points
on: Dispel your justification-monkey with a “HWA!”
I think I more or less try to live my life along the lines of HWA, and it seems to go well for me, but I wonder if that says more about the people I choose to associate with than the inherent goodness of the attitude. HWA works when people are committed to making things go better in the future regardless of whose fault it is. But not everyone thinks that way all the time. Some people haven’t acquired a universal sense of duty, they only feel duty when they attribute blame to themselves, and feel a grudging sense of unfairness if asked to care about fixing something that isn’t their fault. HWA would not work for them, unless they always understood it to mean “other person’s responsibility” and became moral freeloaders.
Even among those better suited to HWA, I think it’s still less than ideal, because it suppresses consensus-building. I think inevitably people will still think about whose fault something is, but once someone utters “HWA,” they won’t share their assessments. When people truly honor the spirit of HWA this won’t matter, because they won’t ascribe much significance to guilt and innocence, but the stories we tell about our lives are structured around the institutions of guilt and innocence, and by imposing a barrier to sharing our stories with one another, we come to each live our own story, which I fear is what tears communities and societies apart. HWA may be good for friendships, but I’m not sure it’s good on larger scales of human interactions.

adrusi 28 Jan 2018 9:17 UTC
8 points
in reply to: cousin_it’s comment on: Pareto improvements are rarer than they seem
I think philh is using it in the first way you described, just while honoring the fact that potential future deals factor into how desirable a deal is for each party. We do this implicitly all the time when money is involved: coming away from a deal with more money is desirable only because that money makes the expected outcomes of future deals more desirable. That’s intuitive because it’s baked into the concept of money, but the same consideration can apply in different ways.
Acknowledging this, we have to consider the strategic advantages that each party has as assets at play in the deal. These are usually left implicit and not obvious. So in the case of re-opening Platform 3, the party in favor of making the platform accessible has a strategic advantage if no deal is made, but loses that advantage if the proposed deal is made. The proposed deal, therefore, is not a Pareto improvement compared to not making a deal.

adrusi 28 Jan 2018 9:40 UTC
1 point
in reply to: Said Achmiz’s comment on: Dispel your justification-monkey with a “HWA!”
What happens the next time the same thing happens? Am I, Bob, supposed to just “accept reality” no matter how many times Alice messes up and does a thing that harms or inconveniences me, and does Alice owe me absolutely nothing for her mistakes?
If Alice has, to use the phrase I used originally, “aquired a universal sense of duty,” then the hope is that it is less likely for the same thing to happen again. Alice doesn’t need to feel guilty or at fault for the actions, she just acknowledges that the outcome was undesirable, and that she should try to adjust her future behavior in such a way as to make similar situations less likely to arise in the future. Bob, similarly, tries to adjust his future behavior to make similiar situations less likely to arise (for example, by giving Alice a written reminder of what she was supposed to get at the store).
The notion of “fault” is an oversimplification. Both Alice’s and Bob’s behavior contributed to the undesirable outcome, it’s just that Alice’s behavior (misremembering what she was supposed to buy) is socially-agreed to be blameworthy and Bob’s behavior (not giving Alice a written reminder) is socially-agreed to be perfectly OK. We could have different norms, and then the blame might fall on Bob for expecting Alice to remember something without writing it down for her. I think that would be a worse norm, but that’s not important; the norm that we have isn’t optimal because it blinds Bob to the fact that he also has the power to reduce the chance of the bad outcome repeating itself.
HWA addresses this, but not without introducing other flaws. Our norms of guilt and blame are better at compelling people to change their behavior. HWA relies on people caring about and having the motiviation to prevent repeat bad outcomes purely for the sake of preventing repeat bad outcomes. Guilt and blame give people external interest and motivation to do so.

adrusi 28 Jan 2018 9:48 UTC
3 points
on: Notification update and PM fixes
Are there plans to support email notifications? Having to poll the notification tray to check for replies to posts and comments is not ideal.

adrusi 2 Jul 2018 5:19 UTC
5 points
on: Are ethical asymmetries from property rights?
I’m attracted to viewing these moral intuitions as stemming from intuitions about property because the psychological notion of property biologically predates the notion of morality. Territorial behaviors are found in all kinds of different mammals, and prima facie the notion of property seems to be derived from such behaviors. The claim, then, is that during human evolution, moral psychology developed in part by coopting the psychology of territory.

I’m skeptical that anything normative follows from this though.

adrusi 7 Jul 2018 19:16 UTC
12 points
in reply to: ChristianKl’s comment on: Debt is an Anti-investment
I think there’s some ambiguity in your phrasing and that might explain gjm’s disagreement:

You seem to value the (psychological factor of having debt) at zero.

Or

You seem to value the psychological factor of (having debt at zero).

These two ways of parsing it have opposite meanings. I think you mean the former but I initially read it as the latter, and reading gjm’s initial comment, I think they also read it as the latter.

adrusi 25 Apr 2019 6:40 UTC
23 points
in reply to: ESRogs’s comment on: The Principle of Predicted Improvement
I also had trouble with the notation. Here’s how I’ve come to understand it:
Suppose I want to know whether the first person to drive a car was wearing shoes, just socks, or no footwear at all when they did so. I don’t know what the truth is, so I represent it with a random variable $H$ , which could be any of “the driver wore shoes,” “the driver wore socks” or “the driver was barefoot.”
This means that $P (H)$ is a random variable equal to the probability I assign to the true hypothesis (it’s random because I don’t know which hypothesis is true). It’s distinct from $P (H = h_{i})$ and $P (h_{i})$ which are both the same constant, non-random value, namely the credence I have in the specific hypothesis $h_{i}$ (i.e. “the driver wore shoes”).
( $P (H = h_{i})$ is roughly “the credence I have that ‘the driver wore shoes’ is true,” while $P (h_{i})$ is “the credence I have that the driver wore shoes,” so they’re equal, and semantically equivalent if you’re a deflationist about truth)
Now suppose I find the driver’s great-great-granddaughter on Discord, and I ask her what she thinks her great-great-grandfather wore on his feet when he drove the car for the first time. I don’t know what her response will be, so I denote it with the random variable $D$ . Then $P (H | D)$ is the credence I assign to the correct hypothesis after I hear whatever she has to say.
So $E (P (H = h_{i} | D)) = P (H = h_{i})$ is equivalent to $E (P (h_{i} | D)) = P (h_{i})$ and means “I shouldn’t expect my credence in ‘the driver wore shoes’ to change after I hear the great-great-granddaughter’s response,” while $E (P (H | D)) \geq E (P (H))$ means “I should expect my credence in whatever is the correct hypothesis about the driver’s footwear to increase when I get the great-great-granddaughter’s response.”
I think there are two sources of confusion here. First, $H$ was not explicitly defined as “the true hypothesis” in the article. I had to infer that from the English translation of the inequality,
In English the theorem says that the probability we should expect to assign to the true value of H after observing the true value of D is greater than or equal to the expected probability we assign to the true value of H before observing the value of D,
and confirm with the author in private. Second, I remember seeing my probability theory professor use sloppy shorthand, and I initially interpreted $P (H)$ as a sloppy shorthand for $P (H = h_{i})$ . Neither of these would have been a problem if I were more familiar with this area of study, but many people are less familiar than I am.

adrusi 10 Mar 2021 5:46 UTC
17 points
on: Deflationism isn’t the solution to philosophy’s woes
I think an important piece that’s missing here is that LW simply assumes that certain answers to important questions are correct. It’s not just that there are social norms that say it’s OK to dismiss ideas as stupid if you think they’re stupid, it’s that there’s a rough consensus on which ideas are stupid.

LW has a widespread consensus on bayseian epistemology, physicalist metaphysics and consequentialist ethics (not an exhaustive list). And it has good reasons for favoring these positions, but I don’t think LW has great responses to all the arguments against these positions. Neither do the alternative positions have great responses to counterarguments from the LW-favored positions.

Analytic philosophy in the academy is stuck with a mess of incompatible views, and philosophers only occasionally succeed in organizing themselves into clusters that share answers to a wide range of fundamental questions.

And they have another problem stemming from the incentives in publishing. Since academic philosophers want citations, there’s an advantage to making arguments that don’t rely on particular answers to questions where there isn’t widespread agreement. Philosophers of science will often avoid invoking causation, for instance, since not everyone believes in it. It takes more work to argue in that fashion, and it constrains what sorts of conclusions you can arrive at.

The obvious pitfalls of organizing around a consensus on the answers to unsolved problems are obvious.

adrusi

One-Con­scious­ness Universe

One-Consciousness Universe