Discussion [Rationality] An Introduction

Well it is with collaboration from Robin Hanson.

The book I reference: Rationality: From AI to Zombies is PUBLISHED.

The book is also nothing more than a compressed compilation of the blog.

The work doesn't require interpretation. It's a simple introduction to Bayesian rationality.

Anyone can understand it if they read for themselves.

I merely sought to "spread the gospel" to NUF.

I got insulted and attacked.

My posts are a reproduction of Eliezer Yudkowsky's work with my understanding and ideas.

The only part I've added some far is "Something I like to call winning. It let's you win".

I think Yudkowsky might have said that somewhere, but it's irrelevant, as I was basing it on my own understanding.

I type on a phone with bad network.

Scrolling up everytime to get more to quote is stressful, and I may end up losing the entire post.

It'll be very inconvenient for me to post the way you wish I did.


I didn't doubt the veracity of his arguments(I don't know about cognitive sciences, but him saying I was saying Bullshit is wrong. As my words are the words of the person commonly identified with the field for now).

Bayesian Rationality isn't something you can get a degree on for now. So relying only on published work(Book is published BTW), is misleading.

 

I said I know Rene Descartes.

 

Epistemological Bayesian Rationality yes.

Instrumental is not so much based on epistemology?

It's more about achieving your goals. It incorporates decision theory.

 

@Aca1814ina - My apologies. I tip my hat to you ma'am.

Pfft.
I like to have a bit of fun when I post things on the internet. I'm not allowed to say things like that in real life because everyone is above medical students in the hierarchy. Even the non-medical staff outranks me. I'm also not allowed to make fun of patients, regardless of stupid they are. So I just have a stream of rambling insults running around in my head all the time, and some of them escape.

Its not the use of a blog that I object to because this isn't academia. You just cannot use a single source as the basis for all of your claims, especially when that source is built on hundreds of years of research and theories by many respected philosophers and scientists. If he actually understood the context surrounding Bayesian Rationality, I'd be less critical.

? What's with the academic-speak? Technically, his claim is only true because interpretation is subjective and I can't actually invalidate something that is unique to him. Any interpretation, regardless of how misapplied or misunderstood it is, is always valid. That middle ground is such a cop out though, and you only use it when you get tired of trying to reason with someone that doesn't know better.

 

Not really. Eliezer Yudkowsky defines instrumental rationality using values.

I actually quoted his definitions of the term verbatim.

My understanding of the term, is that it lets you achieve your goals, values, etc. It lets you systematically and consistently achieve that which you desire.

I am a Lesswronger, and subscribe to Eliezer Yudkowsky's Philosophy.


See, it may not be a matter of right or wrong, but rather. What I'm trying to teach may not be what you associate with the term "Bayesian Rationality".

If you have another alternative concept stored in the folder "Bayesian Rationality", then it's too bad.

I not interested in switching to your concept, and your mother interested in switching to mine.

Let's agree to go our separate ways.

 

Um, now you're running into my areas of understanding. And getting it wrong.

The study of reality and being is specifically metaphysics, which is only a small part of philosophy. Cogito ergo sum is epistomology, a completely separate field about the study of knowledge. Maybe I'm misunderstanding and you meant to make that distinction, but right now the way you've phrased it is incorrect. Also, I'm going to specifically deny philosophy as the science of anything. Science is a study which uses evidence and experiment; philosophy does not.

See, now you're getting it right. Except, of course, you've group beliefs in with knowledge and that's technically axiology which is a different branch. The Bayesian rationality seems to be specifically a form of epistomology, thus beliefs are not really relevant to its practice. (Knowledge vs. beliefs is a distinction I think everyone could understand better, especially myself).

Those are axiological moral theories. Morality is almost exclusively the realm of axiology. Don't bring morality into a debate on epistomology, because that doesn't improve the discussion at all. However, that is all true.

I don't know this part very well. I think this is from Kant, but again I don't know well enough in this area. However, it really seems like you're mixing morality and epistomology, and those do not go together.

And like that I'm lost. Your point was not clear enough to be understood and seemed to be damaged by your prior mixing of different topics. But then again, I dabble only in metaphysics. Morality and epistology go over my head.

Rationality is by definition epistomology. So is decision theory. The reasons that go into that rationality tend to be axiological, though.

 

My interpretation is Yudkowsky's interpretation, and my definitions were his.

It's not a matter of not properly understanding Yudkowsky.

His words were quoted VERBATIM.

You clearly haven't bothered to look up the blogs, or the book. If you had, you'll know that the image you paint of me is wrong.


Until you at least take the effort to correct your misguided image, I won't be interested in discourse; you can't fix a problem when you don't know what it is.

The problem certainly isn't misunderstaning Yudkowsky as you claim(you haven't even bothered to check what he did say, and have been unknowingly disparaging him), so what is it.

@Aca1814ina, no I don't think of Yudkowsky as God. What I'm doing is equivalent to citing Einstein when discussing General relativity. (Not saying Yudkowsky founded Bayesian Rationality), but it's his "brand", that I'm teaching.

 

I'm not old enough for a ma'am. Also not married. But still, better. And yes, I understand the desire to act out on the Internet. And that my manners are considered outdated.

True, but you can't speak for the knowledge of another, simply their presentation of it. It makes you look like a child as well.

And yes, I am acting like a child too. The difference is I know it

Because for some reason I like to try and mediate conflicts that I am not actively involved in. Also, because popcorn.

 

I said his definition was wrong, and linked him to a bite sized definition on www.reddit.com/r/philosophy ; he clearly didn't follow the link .

 

I quoted Yudkowsky verbatim.

It is extremely improbable that I misrepresented Yudkowsky. Not only due to quoting him verbatim, but because his book is my Bible; I read it daily, and follow it judiciously.

 

I'm going to let you in on a little secret from my theorist friends: if you only quote Einstein, you don't get published.

Even if you have the most groundbreaking work on general relativity since he did. You have to quote many, many sources from all the people before and since who helped him build up said theory.

And how often do people misinterpret the bible?

Actually, that's a religious argument. Ignore it, please.

 

Instrumental Rationality is epistemological? I never did think of it that way.

It deals with achieving objectives?

Would you classify that as epistemological. If so, please explain.

 

I quoted him verbatim.

@Devshard hasn't read Yudkowsky's works.

He was attacking me while praising Yudkowsky. If he knew I was quoting Yudkowsky, he wouldn't do that.

He praised Yudkowsky to the sky, while indirectly saying he teaches bullshit.

Yudkowsky provides several links in his writing.
(My phone had over a 100 tabs opened), but the only links I've read are the links he's provided in his writing. Even the other books on Rationality that I'm looking for free e-books of are only those linked hey Yudkowsky.

So my knowledge is exclusively filtered through him.


I know his brand of Rationality, and it is that I seek to teach.

@Devshard, and I have a disagreement, that I'm not sure where it originated.

None of the allegations he levies against me are true.


I follow and understand well Yudkowsky's dogma. That is what I seek to teach.

Tell me Milady; have I ever stated otherwise?

 

I don't think we're necessarily talking about the same thing.

It maybe what I and Yudkowsky call "Bayesian Rationality" differs from us.

Take a look at www.lesswrong.com

Are we speaking about the same thing?

After you've done that, and APOLOGISED

For calling me a "Deluded HPMOR reader" among other things, then we can resume cordial conversation.

 

Manners aren't outdated.

"Do unto others as they will do unto you"

I respond to civility with civility, and to hostility in kind.

 

The "Cogito ergo sum" essay is used mostly for epistemology, yes. But it also used for cognitive theories, symbolic interface theories, and reality-interaction work. Our neural architecture and functions place our 'mind' on a different plane of reality. The best illustration of this is how we use memory to violate the time-space continuum. Nothing in our universe can return to a previous point of time in a given space because time is a fundamentally positive vector. Using our memories, we can technically return to a previous time while our bodies occupy a given space which violates the positive integer attribute of time. That's a planar separation of reality.

Specialization in fields of study is a relatively new thing. Newton, Liebnitz, Bernoulli and all of those guys identified themselves as philosophers. Not as physicists, mathematicians, or scientists. Philosophy always uses the scientific method and is completely observation and evidence based. I'm talking about formal academic philosophy, not the armchair/reddit variety of philosophy.

Sigh, no. There isn't as distinct a separation between them as you're thinking. Bayesian rationality involves both belief and knowledge because we use both in any cognitive reasoning circuit. The Bayesian theory is specifically about predicting future decisions and examining the knowledge and beliefs involved in that reasoning circuit. The neural architecture of the human brain simply cannot make a decision without using both moral reasoning and absolute knowledge.

Actually, that's not true. There is a variant neural architecture that people with anti-social personality disorder have that can make decisions without firing the moral reasoning circuit, but that's just because they don't have one or it doesn't function. (Anti-social personality disorder is more commonly known as sociopathy or psychopathy.)

The framework behind those two can be applied to axiological moral theories as well as decision making. Look up Josh Greene's fMRI work from Harvard.

Let me try again. Bayesian rationality is calculating the probabilities for a series of future actions/decisions so you can calculate the probability of them leading to a defined outcome in the future, and then examining/altering the underpinnings of each of the individual future actions/decisions to increase the probability of the defined future outcome being true.

So: P(FutureOutcome) = P(FutureEventA) + P(FutureEventB) + P(FutureEventC) + P(FutureEventD)

Now, when you make modifications to increase the probability of FutureEventA occuring, you might also do something that decreases the probability of FutureEventC or FutureEventD occuring or both. You basically need to use careful introspection and detailed analysis to increase your individual future event probabilities while also accounting for loses and preventing huge drops.

Justification is entirely decision theory, not axiology.

 

I find this statement EXTREMELY doubtful.

Rene Descartes for example identified with both. Both as a philosopher and Scientist/Mathematician.

Reddit Philosophy isn't armchair though? Academic Philosophers participate on Reddit. I've had the privilege to engage Masters in Philosophy through Reddit. I know at least one Professor on the subject who also participates there.

 

I find this funny. ASPD isn't always born. It can easily be made.

I can make decisions without morals though (my morals aren't really moral in the first place)? I'm amoral to an extent. I know a Sociopath.

If ASPD is made not born, then how do you suppose those with it lose the neural circuitry to invoke moral reasoning.

Amoralism is actually a Philosophy. Check it out.

 

And here we got a person claiming as if a theory equal truth statement. I was late in posting this since I was asleep, however, a theory is still a subject for debate. Anyone who claims that they're smarter just because they are knowledgable on few books should know this theory restriction, that is, if you are truly smart. I call, Order upon this argument. And let me offer you a few statement, "Narrowminded Philosophy STUDENT has never been called smart from an outsider's perspective." Try completing my EDBO (Eight Dimensional Box Operations). You should pioneer some stuff while your at it if you want your opinion heared. Ah, I'm currently on my phone by the way. -A certain psychology undergrad determined, and told by professors, to make a book about the EDBO. I know less than you but I am a known genius (I mean called by more than you), capishe: @Drevwhatevs

 

Okay, the key point behind instrumental reasoning is the interaction with other people and the environment. The kind of goal is irrelevant. The only important point is that everything is done ONLY to achieve that goal. The problem with using instrumental reasoning to achieve something axiological in nature is that you create a paradox in which all of your actions and interactions with people depend on the outcome of gaining that value, but that process makes it impossible for you to actually gain said value.

Gaining something intangible like a value is a continuous process of self-improvement. There is no point where you'll ever achieve success in that goal, so it cannot be compatible with instrumental reasoning.

Again, this is a part of everyone's cognitive architecture already. It only works for short-term goals with a defined process. The instrumental reasoning circuitry is incompatible with long-term or distant future goals and also incompatible with intangible things. There is no amount of training or thinking that can allow you to use that cognitive architecture for those things. The brain does not allow for it to happen.

Hush. I've known and respected Yudkowsky since you were 8 years old. I've actually met the man in real life, and I've gotten his opinion on a few of my papers as I was writing them. You want to claim that you understand his work better than I do because you read some of his blog posts and a book written in laymen's terms?

I have a copy of that book also, and I've read it more than once. You didn't quote Yudkowsky nor did you include the proper context or explanation from that passage.

Saying the words "Yudkowsky's dogma" goes completely against everything Yudkowsky stands for as a philosopher, scientist, teacher and human being. He's almost a pure skeptic and literally screams at people to question everything they hear/read/see over and over again.