Tiled With Pentagons

(Epistemic status: Neat-seeming ideas that promise to start paradigmatizing agent foundations, as expounded by JSW... but I dunno, pal, I'm just some research-creature. I didn't run this by him and I didn't consult notes or videos, either, so it might be very wrong or desperately incomplete in places. But also: in neglecting/refusing/failing/setting-off without recourse to notes, maybe I'll say something new and worth poking at some more.)   In seeking a paradigm for AI safety or AI alignment, we sometimes find ourselves seeking a paradigm for agent foundations, the study of what mind-type things that take actions for reasons towards goals might do in maximal generality. But where do we even begin? We find solace in materialistic reductionism, shunning most of metaphysics in the process: agents are a phenomenon first and foremost of the world of things, so it's as a very special type of thing that we will try to understand them.   From reductionism we take a guiding...

(Epistemic status: A creased, stained map to what were once my favorite hunting grounds. Accessible to anyone who can support substantial abstraction; prior math knowledge is not necessary. In particular, ignorance of calculus is not an obstruction here, but total ignorance of geometry or like, arithmetic or logic, will be. Extremely dense and probably won’t get you there, but at least you’ll ask better questions. Partially dedicated to DG, JM, and PR.)   The path to my old hunting grounds (as pictured above) is long and twisty and winds through a lot of necessary  math along the way. You'll be best served by carefully going through the parentheticals and answering the questions I ask, so as to keep track of the blazes and to keep your footing. You'll still get something out of this if you move more quickly, but you might end up lost further down the line. You might find the pace a little slow if you already know the territory - test yourself by answering those same par...

(With apologies to the Seattle Garages.) Much has been made of two major ways of relating to personal identity, neither one better than the other, and both of them seemingly innate. There's episodic identity, where you consider yourself a different person altogether from who you were in past years, and diachronic identity, where you consider yourself to be the very same person in all respects to who you were long ago. Naturally, there are in-between ways to see yourself: the timescale of episodes can vary widely, trending towards diachronicity at the long end; alternately, the starkness of the episodic cuts can be softened, with each new self being a new version rather than a complete renewal. For myself, I tend towards the diachronic side, but not completely. When I think about the course of my life to date, I tend to divide it into rough eras or arcs; I've been the same person throughout, with mostly the same core drives and motivations and capacities and interests, but have ...

Here are a few more delicious and mostly scalable recipes that I’ve made repeatedly with great results which I mostly haven't committed to text. Unfortunately, it's still true that none of them are vegan, though the desserts at least are vegetarian. This is a recipe blog now, I guess. (It’s not. Don’t worry. Or do; I’m not your parent.) No pictures this time, as I'm travelling; maybe I'll add some later. Roman Showup Stew: An excellent lentil-based stew of the kind that ancient Romans might recognize, except for the part where it's got a bunch of unfamiliar but superior ingredients in it, was cooked using exceptionally fine tools, has a few absurdly expensive (for the time) spices and ingredients in it, and also it'd be too spicy for them to eat. Take that! 1 can lentil soup (I favor Progresso) 3 large slices of Spam or equivalent mass of bacon (optional but highly recommended) 1/2 a large red onion 8-12 cloves garlic  1 large russet potato (opt) 1 medium carrot...

(Epistemic status: Personal experience and observation, if well-attested. Associated advice.) We live in an age of ludicrous material plenty, but all too often, people seem to find themselves lacking something valuable or delicious about life. What good is that sort of notional wealth if your sensory experiences don't seem to reflect that plenty? Clearly, something is missing in the way many people live. if they find themselves dissatisfied with the texture of their daily lives amidst such wealth, unable even to figure out what it is they're missing. What could possibly bridge that gap? The answer to that riddle is in the very same observation. Consider that many little luxuries nowadays would have been absurdly expensive or even bluntly impossible to come by in ages past, perhaps even as recently as the 18th century. Given the way in which desire and assessment of value often change slowly over the centuries, one can reap massive phenomenological gains by finding that which wo...

(Epistemic status: Avoiding cashing this out too concretely to avoid drawing fire... but let's do ourselves a service and refuse to pretend to see symmetry where it isn't, shall we?) In the political sphere, it's been said of late that every accusation is an admission. If Mr. Richard Reprobate of the Purple Tribe levels the accusation of orphan-shredding at Ms. Iamb Innocentia of the Oranges, or if he fearmongers about the dread furnaces which even now the Oranges are cleaning out and stoking to roast the poor Purples in, it will with depressing frequency be the case that our dear friend Dick is putting in an order for new rotary blades, or that the Purples are privately bikeshedding about the relative merits of rotisseries as opposed to applewood smoking or pit barbecues. More generally, whenever some stuffed shirt, public figure, or even just someone you know claims that their adversaries will take such and such an aggressive action, or that they engage in such and such h...

(Epistemic status: Yet another post that I hope dearly that you only dimly comprehend, but if you understand it all to well, this is for you. In particular, for LW.) There is a place beyond sadness, beyond grief, beyond despair. In the lightless depths of that place, there is no meaning, no joy, and precious little sensation. Food turns to ash in the mouth, and the body feels wooden and alien, and living each day feels like an unbearable burden. If you find yourself there, know that I sometimes find myself there, too. One thing to internalize: you must lose hope, and endure all the same. To those who come again and again to that abyss, to hope is to stare into the sun and be blinded by it; to desire is to hold fast to thorns and bleed. So don't do that, not for now, not until you drift back up or claw your way out. Just be stable and keep space for that day on which you can live again. Gather your energy; scrape it together and stopper it tightly inside yourself against the corrosi...

(Epistemic status: A linguistic study. Still not real, but also still the kind of not-real that's real.) [Language] has a rich set of evidentials to express how a speaker came by a piece of information, and while the use of evidentials isn't technically obligatory, their lack is notable and at least a little suspicious. While like other languages [Language] marks evidentiality through an affix to the verb or stative, unlike most evidential-bearing languages like Turkish, Aymara, Dagestanian, Tibetan, and Bulgarian, [Language] has numerous evidential classes, categorizes evidentiality differently, combines mirativity with a couple of the evidential classes, and permits the use of explicit numerical parameters in evidentials when appropriate. Its use of evidentials is closest to Lojban's use, though the resemblance is only moderate. We list the most commonly-used ones. 1. Direct perception. This can be sight, hearing, smell, or any other variety of direct sensory perception;...

(Epistemic status: hard-won personal experience; advice. Requested by PR and LW.) How to Move Cross-Country in the US Without Losing Your Stuff or Going Insane First, come to terms with the fact that this is going to be fairly expensive. Even if you have relatively few belongings, and don't have things like a large desk or bed to bring with you, you're still likely to spend upwards of US$1,000. The next thing to do is pack up all your belongings. If there's anything you plan to discard or give away, do that at some point during the process. For everything else, you'll want to go to a moving store like U-Haul both for boxes and for assorted packing materials, like newsprint, bubble wrap, dish boxes, and especially packing tape - get one more roll of it than you expect to need. You may also want to get colored tape, to color-code box priority; I recommend categories for low-priority, mid-priority, high-priority, and top-priority, in terms of when you should unpack the box...

(Epistemic status: The butterflyest of butterfly ideas. Buried lore that I dug up from a handful of places; pieces that pretty much only I would think to put together, if pieces they are. I'm thus writing this up so it doesn't get lost. This might be incomprehensible or otherwise not hold together too well... but this smells like there's something here to fit together or find. Heck if I quite know what.) (With thanks to MB, JSW, MM, and CRW.) Quick overview: We have grounds to believe that decision theories (like (PO)MDPs) that also use infrakernels should then be able to converge to optimal policies on the kind of weird decision-theoretic puzzle (for instance, Newcomb) where causality doesn't seem to quite work right or where extra information can hurt you: specifically, there's a somewhat obscure connection between Markov categories, information theory, and probability theories where in defiance of Kolmogorov, probabilities can add up to less than 1. Background Fa...

(Epistemic status: Something I wish I'd had when I was a tiny undergraduate and then a yearling graduate creature, so I'm doing it for them now. I've been in the trenches, my soul's made partially of math at this point, and I helped PR - who this is partially dedicated to - do an algebra pset this week about using the Sylow theorems. Nevertheless it's been a while and I was never a dedicated algebraist; in particular, some of the "secret tech" arguments at the end have a very geometric-group-theorist flavor. If I got anything wrong I want to know. This will be useless to you unless you know a decent amount of abstract algebra already.) A major kind of problem that shows up often for abstract algebra qualifying exams is called a "Sylow problem". They're generally of the form "Show that if \(|G| = 69\), it contains a normal subgroup" or "Classify all groups of order \(G = 42\)", and they rely on the Sylow theorems. This isn...