YoungAchamian
No bio...
User ID: 680
YoungAchamian
6hr ago
I mean if “member in good standing with community” is your barometer then absolutely what i said applies to prog spaces. I exist in prog spaces and the avg white man is absolutely accepted as long as they mutter the shibboleths.
If anything they don’t really require you to believe them so much as you don’t cause waves or have some plausible deniability.
Now the infighting you are alluding to is among the leadership, the narcissists, the sociopaths. Aka not normal folks. And yes to have power requires special circumstances in the memeplex, but hereditary or ideological purity as a metric of power inside a movement is not unique.
YoungAchamian
7hr ago
The wokes are in the Proselytizing phase of their religion, akin to islam. Christianity had that phase but is post-Proselytizing now
YoungAchamian
7hr ago
It’s a path to redemption not a path to place of honor. Unless you would claim that Christianity honors equality all sinners who have joined in following Christ. But considering the hierarchical nature of most churches I think thats a lot to swallow
Also, creepy megachurch pastors? Pedo Priests? Allowing people in power to abuse that power is classic human dynamics. Cast not the first stone unless ye are free from sin.
YoungAchamian
9hr ago
Sure it does, be a good male ally, don’t take up space, defer to women, acknowledge your privilege, pay for your sins, accept any abuse hurtled at you as remuneration for your sin.
Its all very Christian in nature.
YoungAchamian
12hr ago
These fall into the RL side, specifically Model-based RL. The dirty open secret about model based RL is that the RL-policies(ml-models) trained are only as good as the sim-models (there are two types of "models" here that mean different things, smh...), they don't generalize well. So for the RL trained on Starcraft, it is also then tested on Starcraft, so good training results in good testing results. The major leap comes when you transfer it to a real world scenario, where it will not perform, because Starcraft is not a good model of the real world. Part of this is strictly technical, RL models/policies learn a mapping of states to actions, ie which action to take in a given state for the best reward now or in the future. The Starcraft world actions don't exist in reality. The models don't really learn high level tactics that can be extracted and re-used (unfortunately)
Modeling the real world with finite states is probably impossible. Near infinite/continuous states is going to be insanely hard to model, so if they are doing this, then good luck but it's a likely dead end.
YoungAchamian
22hr ago
behind AI simulations of potential Taiwan scenarios
Not to derail, but does anyone have any idea of what this actually means? I work with defense modeling and sims folks and I am an AI/ML person but whenever I hear the words AI Simulations I'm left scratching my head. Is this just making world sims for RL models? But then why would you be training an RL model to invade Taiwan? Simulations are generally done on big HPC rigs with tons of variables, ie. Finite Element Modeling. This isn't something that really benefits from neural nets. Maybe my technical classification is different than the non-technical, but this guy is supposedly some researcher I feel like the classification would be more precise.
YoungAchamian
1d ago
Yeah... it sucks that its going to have to be such a costly lesson but idk how else to get them the message at this point. I assume their careers and self interest are super aligned with foreign adventures so to stop this kind of behavior will require a steep cost to them personally and their careers, to act as a warning for the next gen of war hawks.
I also want Congress to rein in some of their powers. I'm tied of "special operations" that take a whole carrier battle group over months. If you are doing large scale operations, you are at war.
YoungAchamian
1d ago
I'm a pretty cynical bloke, and my sympathy for Trump didn't even come from his promises, but from the kinds of people who hated him,
Pretty much my stance when I voted. Trumps not really my "hated-outgroup" even if I'm not a fan. but the kind of people who hated him are. And I think he's a vindictive bully who would attack my hated-outgroup, dismantle their hold over American institutions. Until Iran it was kinda looking that way., now... idk. But I am not happy.
YoungAchamian
1d ago
What does the anti-war side in the US want in the Iran conflict?
A Domestic policy focus, the continued removal of the woke cancer from American institutions, a weakening of the war hawks and the globohomos, a reduction in executive fiat and the abrogation of congressional powers and at this point a complete divestment from the sandbox from hell.
The War Hawks keep seem to be either missing the message in regards to the average citizen’s desire for empire building in 3rd world shitboxes, or more realistically they understand if they get us into a quagmire then with no good options to extricate ourselves without humiliation they can get their way. They keep hitting the defect button, at this point we might as well make the cost for doing so, ugly otherwise they will never get the message.
YoungAchamian
2d ago
There's always more tax cattle
The standard term is pay pig /s
YoungAchamian
3d ago
if you don't even read what they say, you cannot be considered to be knowledgeable about their thesis.
They broadcast the hell out of their thesis online, as far as I have encountered it, I categorically disagree with the base principles.
Keep in mind, making these arguments is akin to telling early NASA engineers in the 1950s that they shouldn't send rockets in to space because then the hyper advanced aliens in the dark forest are going to start collapsing space and that will be bad for the universe. It comes off as literal insanity. Just because some science fiction book has conceived of it doesn't make it actually possible or even reality.
I'm tired and I'll respond to the rest later.
YoungAchamian
3d ago
This jives with my understanding, I always have thought of them as similar to a software consulting company for Defense related projects.
YoungAchamian
4d ago
Have you read Superintelligence: Paths, Dangers, Strategies? That's a pretty core doomer text.
No I haven't and I'm not likely to. The fun of Science fiction is its not taking itself too seriously.
AIs naturally want
Let me stop you here, AIs want nothing, they aren't sentient. They are a very advanced token model that is predicting the desired output from the context and the question. They are a tool, a mathematical function approximation fitted to a general solution. If doomers want to call a coding subroutine sentient, well its a free country, but they are abusing the english language to do it. This is the cargo-cultism.
let it give instructions for a civ 4 game
In AI R&D we call this Course of Action Generation. The military has been trying to get GenAI to provide strategy tips for the better part of 4-6 years. It has failed every wargame it has attempted. If you think you have a great solution and that AI can totally model the world for military tactics, I recommend you submit to this: https://sam.gov/workspace/contract/opp/60a94bf650a84d3fb0bb524862e78401/view, DARPA's DISCORD project. If DARPA is asking for it, that means they think it doesn't already exist and is a moonshot.
Is that not a world model
Not if you are filtering through yourself to give it context and understanding of the situation. You are using your human world model as the surrogate to the LLMs. Can it play Civ 4 left to its own devices, shown a picture of the screen with the tutorial on? Maybe an agentic setup to take actions?
Saying that AIs don't have a world model today is not an effective counter
To be specific, I said LLMs don't have a world model. LLMs are not the full set of AI... Do I believe AGI will never develop a world model? No.
The whole concept of a world model is nebulous
This is like saying a convolution or self attention is nebulous. World Model's mean very specific things technically in ML research. The charitable interpretation is that converting the technical jargon to a descriptive lay-person understandable explanation is very challenging.
YoungAchamian
4d ago
Exactly, which is my major gripe with the AI-Science-Cargo-Cult Mysticism that AI singularity doomers swim in. Basing the real-world situation on the details of the Sci-Fi scenario that is not in any way based on actual science is insanity. It's there for broad outlines, core elements, not details.
And honestly I haven't seen as much motion as I'd expect on the world interaction front.
It's a very active area of research but it hasn't reached the state that lay-folk would interact with it.
YoungAchamian
4d ago
A "world model" is not really some far reaching philosophical characteristic. LLM discussions require you keep in mind what the LLM is actually doing. It is making very impressive statistical correlations between the semantic mapping of token embeddings in a way that best conforms to the expected output. It is not modeling anything else than that. Whatever world knowledge it has is only acquired indirectly through patterns in data.
If I asked you why does a ball fall if I drop it, you'd say gravity, if I dropped the ball you'd expect it to fall. If I asked why does my glass of water have less water in it after an hour outside in the sun, you'd say evaporation. If I showed you a small metal projectile moving at high speeds towards a person, you'd realize they are being shot at, that blood will come from the wound, that they will give a cry of pain, that someone was doing the shooting, that there should be a loud noise, etc. You have a "model of the world" aka you understand that there are causal factors that will cause a reaction and that if you observe the reaction what those causal factors may be. These are all things that happen irrespective of the statistical correlation of words.
Now, Obviously an LLM will be able to tell you all these things because all of these situations occur as text in its training data, over and over. But there exists things that exist outside of that training data, and so you should expect it to act very much this same way for those.
YoungAchamian
4d ago
Same with the passenger. Sucks to be him because I'm fine.
I mean in this case the passenger is analogous to the baby.
I feel this topic will endlessly fail to converge because of the underlying teleological differences. If one side believes in a unitary telos of sex -> reproduction and the other believes in a multi-variate telos of sex, then I think convergence is unlikely. The risk-based argument fundamentally belongs to the multi-variate telos. The unitary telos smuggles in Christian values, which is not very convincing to non-Christians. There are other arguments, but ones that are anchored on the topic of life/humanness are always semantic debates where each side trots out their specific very biased tailored definition and attacks the opposing definition.
Unfortunately many of both sides are also invertebrate hypocrites around the principles of this discussion, and want to cut a fine boundary carveout for their pet side while dodging the "child-support payments" of the downstream costs. The best that can we can really hope for is that each side sticks to holding their own views in their own community and doesn't try to enforce it on the other side's community. But due to the universalist and totalizing natures of the underlying grand narratives involved that sort of equilibrium is probably not emergent.
YoungAchamian
4d ago
Your solution is essentially what Chain-of-thought prompting is doing. In order to cut down on hallucinations, aka token prediction unmoored by the constraints of reality, you provide the model the intermediate steps at each section. As you found, when given micro steps, the LLM doesn't need to make large reasoning jumps that require intrinsic understanding of the rules, causal factors or some underlying structure of the world.
Your idea is essentially, "how do I provide an embedding that represents the state of the world" at each step of the model's prompting. Language is the easiest for human's to embed but it is not the most compact or highly representative.
YoungAchamian
7d ago
This is a non sequitur, we are discussing the meta level comparison between pro-life and forcing vaccination, if you wish to discuss bodily autonomy applied to abortion there are comments throughout the thread discussing it.
YoungAchamian
7d ago
ZanarkandAbesFan
Nah ^ blocked me for this comment. Or at least has very recently blocked me and it just so happens to coincide with their low effort comment below. I just find it funny or ironic of all my controversial comments that the genuinely innocuous one was what got me.
YoungAchamian
7d ago
Is it? That's cool. They did not offer me one when I went to a Hillel in college for an antro of religion observation but it also wasn't a very serious service/community.
YoungAchamian
7d ago
In a data modeling understanding, everything is a "piece of evidence" any observation, any data, is fuel for the bayesian model. I observed it, made a tentative connection, stated my theorized connection, you and the other poster pointed out why is not a likely connection, I updated my understanding model of the what/when/where/why's. The end.
Idk why you are jumping at this so aggressively.
YoungAchamian
7d ago
Yes the non-tongue and cheek answer is that this is a problem that has been semi-solved by having a more egalitarian relationship with both partners working at a comparable income. Assortative mating within your socio-economic class leads to stable relationships. I'm not sure where IraqVeteran8888 falls into the spectrum, but my shot in the dark is that a guntuber is conservative -> trad relationship -> Wife stayed home to take care of the kids and do the home making -> bigger alimony payment when divorce happens.
The other observation, though probably more charged, is that average gender warriors opinion of women getting the short end of the stick through their own actions in the dating, is to say "you should have picked better, or actually cared about the red flags". Unfortunately that same advice applies to men in the marriage market. Don't marry Stacy because she's hot and wants to raise your kids if she also is the type of vindictive or mercenary to take you to the cleaners in the divorce.
YoungAchamian
8d ago
I have no dog in this fight, nor am I particularly Joo-pilled. It was more of a curious piece of evidence that might have weighed in on a model of behavior. Not every underlying model mechanism is correct.
YoungAchamian
8d ago
And other predators still, like Christianity must be fought off with memetic technology like rationalism, critical thinking, and atheism. However it was a clever predator and it to adapted to the anti-bodies and now gave us a virulent strain of Atheist-Christianity aka Woke-ism.
No different than if your priest tells you to shut up and listen because you are a sinner… I’m not Christian but if your local priest decides you cant partake in eucharist, as punishment for certain actions, in certain denominations aren’t you no longer in good standing?
Yeah, if i don’t want to say grace or engage in situations that require me to affirm Jesus, that would be analogous. The belief that religious communities don’t police shibboleths is incorrect.
Honestly the only real shibboleth i have really encountered is pronouns. Nobody has asked me to wax feminine balls
More options
Context Copy link