NaiveTortoise

• NaiveTortoise 21 Mar 2018 23:46 UTC

  9 points

  on: The Math Learn­ing Experiment

  Awesome idea! I’m curious about some of the details of the event if you’d be willing to share, partly because I’ve thought of trying something similar for programming stuff:

  • What math topics did you cover and at what level of detail did you cover them?

  • Was the instruction purely informational (one person explaining to another) or were there also problems involved (tutor posing a problem and giving the tutee time to think it through and try to solve it)?

  • Were there people who consider themselves “bad at math” or “not math people” included? Did their views change at all after the event?

• NaiveTortoise 12 Apr 2018 5:08 UTC

  4 points

  on: Two types of mathematician

  Late to commenting on this post, but where would Grigori Perelman, the prover of the Poincare conjecture, fall then? I remember this quote from his biography:

  Golovanov, who studied and occasionally competed alongside Perelman for more than ten years, tagged him as an unambiguous geometer: Perelman had a geometry problem solved in the time it took Golovanov to grasp the question. This was because Golovanov was an algebraist. Sudakov, who spent about six years studying and occasionally competing with Perelman, claimed Perelman reduced every problem to a formula. This, it appears, was because Sudakov was a geometer: his favorite proof of the classic theorem above was an entirely graphical one, requiring no formulas and no language to demonstrate. In other words, each of them was convinced Perelman’s mind was profoundly different from his own. Neither had any hard evidence. Perelman did his thinking almost entirely inside his head, neither writing nor sketching on scrap paper. He did a lot of other things—he hummed, moaned, threw a Ping-Pong ball against the desk, rocked back and forth, knocked out a rhythm on the desk with his pen, rubbed his thighs until his pant legs shone, and then rubbed his hands together—a sign that the solution would now be written down, fully formed. For the rest of his career, even after he chose to work with shapes, he never dazzled colleagues with his geometric imagination, but he almost never failed to impress them with the single-minded precision with which he plowed through problems. His brain seemed to be a universal math compactor, capable of compressing problems to their essence. Club mates eventually dubbed whatever it was he had inside his head the “Perelman stick”—a very large imaginary instrument with which he sat quietly before striking an always-fatal blow.

  from Perfect Rigor.

• NaiveTortoise 11 Jun 2018 16:56 UTC

  5 points

  in reply to: cousin_it’s comment on: The Case Against Ed­u­ca­tion: Why Do Em­ploy­ers Tol­er­ate It?

  Do you think education makes college graduates into standardized employees though? In my experience, watching other standardized employees had more influence on me becoming a standardized employee than going through college.

  I’d also argue that what you described is at least partly signaling. The group of “10 unique autodidacts” will, on average, be less conformist than the group of 10 recent college graduates prior to either group undergoing the experiences that define them in this hypothetical. The fact that they decided to become autodidacts is a signal of that pre-existing tendency, admittedly reinforced by whatever experience had not following the standard track.

  To now argue against myself, I agree with the underlying point that I failed to discuss why hiring often does (and should) optimize for not making mistakes rather than finding optimal individuals.

• NaiveTortoise 12 Jun 2018 2:27 UTC

  3 points

  in reply to: PeterMcCluskey’s comment on: The Case Against Ed­u­ca­tion: Why Do Em­ploy­ers Tol­er­ate It?

  This is a good point, but I’m not sure how to productively integrate it into my discussion of conformity yet. It seems like the higher level point this points to is that conformity exists along many only partially interacting axes (in this example, the type of conformity being an English major shows is different but not *entirely* different from the type of conformity being an Economics major shows).

  Unfortunately, going from a one-dimensional model of less to more conforming to a multi-dimensional one where you can conform along many axes makes this even harder to discuss and investigate… To do so, I suspect I’d need to narrow down my investigation to one industry within which conformity mostly varies along a single dimension.

• NaiveTortoise 12 Jun 2018 2:36 UTC

  4 points

  in reply to: Dagon’s comment on: The Case Against Ed­u­ca­tion: Why Do Em­ploy­ers Tol­er­ate It?

  I agree that this is a larger and more complex puzzle, which is in fact why I avoided addressing it in my post. That said, I want to address some of your points, with which I mostly agree, individually.

  Regarding 1), this is true but, all else equal, you’d think that employers would still have incentives to find cheaper or quicker work-arounds for extracting the same signal (see my discussion of Google hiring directly out of high school in the post). I suspect they don’t because of the other reasons you mention in your comment.

  I agree that 3) and 4) play large roles that are hard to model in economics studies. While I haven’t read it, I wonder if Tyler Cowen’s “The Complacent Class” discusses this.

• NaiveTortoise 12 Aug 2018 18:47 UTC

  1 point

  in reply to: sayan’s comment on: Open Thread Au­gust 2018

  Also interested—programmer with very limited knowledge of RL.

• NaiveTortoise 7 Nov 2018 19:19 UTC

  5 points

  in reply to: megasilverfist’s comment on: No Really, Why Aren’t Ra­tion­al­ists Win­ning?

  Is there an actual description of turbocharging training beyond “deliberate practice but where you think hard about not Goodhart-ing and practicing the wrong thing”?

• NaiveTortoise 9 Dec 2018 19:26 UTC

  LW: 3 AF: 1

  AF

  on: New safety re­search agenda: scal­able agent al­ign­ment via re­ward modeling

  Was anyone else unconvinced/​confused (I was charitably confused, uncharitably unconvinced) by the analogy between recursive task/​agent decomposition and first-order logic in section 3 under the heading “Analogy to Complexity Theory”? I suspect I’m missing something but I don’t see how recursive decomposition is analogous to **alternating** quantifiers?

  It’s obvious that, at the first level, finding an $x$ that satisfies $\varphi \left(x\right)$ is similar to finding the right action, but I don’t see how finding $x$ and $y$ that satisfy $\exists x\forall y\varphi (x,y)$ is similar to $A_2$‘s solving of one of $A_1$’s decomposed tasks is similar to universal quantification.

  To take a very basic example, if I ask an agent to solve a simple problem like, “what is 1+2+3+4?” and the first agent decomposes it into “what is 1+2?”, what “what is 3+4?”, and “what is the result of ‘1+2’ plus the result of ‘3+4’?” (this assumes we have some mechanism of pointing and specifying dependencies like Ought’s working on), what would this look like in the alternating quantifier formulation?

• NaiveTortoise 16 Dec 2018 14:59 UTC

  3 points

  in reply to: abramdemski’s comment on: Multi-agent pre­dic­tive minds and AI alignment

  I’d be very curious to hear more about your general dislike of predictive processing if you’d be willing to share. In particular, I’m curious whether it’s a dislike of predictive processing as an algorithmic model for things like perception or predictive processing/​the free energy principle as a theory of everything for “what humans are doing”.

Bab­ble, Learn­ing, and the Typ­i­cal Mind Fallacy

[Question] How do we iden­tify bot­tle­necks to sci­en­tific and tech­nolog­i­cal progress?

• NaiveTortoise 1 Jan 2019 0:27 UTC

  LW: 6 AF: 3

  AF

  in reply to: janleike’s comment on: New safety re­search agenda: scal­able agent al­ign­ment via re­ward modeling

  Thanks a lot! This definitely clears things up and also highlights the difference between recursive reward modeling and typical amplification/​the expert imitation approach you mentioned.

• NaiveTortoise 2 Jan 2019 19:26 UTC

  4 points

  in reply to: ryan_b’s comment on: How do we iden­tify bot­tle­necks to sci­en­tific and tech­nolog­i­cal progress?

  You should promote this to a full answer rather than a comment! It more than qualifies.

  Regarding 1, I suspect a lot of recent progress in neuroscience has come from applying computational and physics-style approaches to existing problems. See, for example, the success Ed Boyden has had in his lab with applying physics thinking to building better neuroscience tools–optogenetics, expansion microscopy, and most recently implosion fabrication.

  I think nanotechnology is a prime example of 2. AIUI, a lot of the component technologies for at least trying to build nano-assemblers exist but we lack the technology/​institutions/​incentives/​knowledge to engineer them into coherent products and tools.

• NaiveTortoise 2 Feb 2019 23:53 UTC

  3 points

  on: Build­ing up to an In­ter­nal Fam­ily Sys­tems model

  Really enjoyed the post, thanks!

  I started the Earley book and it’s definitely a struggle. I usually can handle “soft skills” books like this one without getting frustrated by the vague, hand-wavy models—I really enjoyed Gendlin’s Focusing, for example—but this one’s been especially hard. That said, having your model in mind while I’m reading has kept me going as I’m using it as a sort of Rosetta’s stone for some of Earley’s claims.

So you want to be a wizard

• NaiveTortoise 31 Mar 2019 18:15 UTC

  2 points

  on: Why Plan­ning is Hard: A Mul­ti­faceted Model

  The discussion of planning across domains seems to ignore the fact that often the best solution to planning a project in a domain with which you’re not familiar is to hire/​get help from another person. Of course, once you turn planning into a multi-person activity, which I think any comprehensive model of planning should treat it as, you also need to factor in uncertainty about others’ plans, which complicates the model quite a bit.

[Question] What are effec­tive strate­gies for miti­gat­ing the im­pact of acute sleep de­pri­va­tion on cog­ni­tion?

• NaiveTortoise 31 Mar 2019 23:31 UTC

  6 points

  on: An1lam’s Short Form Feed

  I’ve recently been obsessing over the idea of trying to “make math more like programming”. I’m not sure if it’s just because I feel fluent at programming and still not very fluent at abstract math or also because programming really does have a feedback loop that you don’t get in math.

  Regardless, based on my reading it seems like there’s a general consensus in math that even the most modern theorem provers, like Lean and Coq, are much less efficient than typical “informal” math reasoning. That said, I wonder if this ignores some of the benefits that programmers get from writing in a formal language, i.e. automatic refactoring tools, fast feedback loops, and code analysis/​search tools. Also, it seems like a sufficiently user-friendly math theorem proving tool could be useful for education. If kids can learn how to program in Javascript, I have to believe they can learn to prove theorems, even if the learning curve’s relatively steep.

  Maybe once I play around with Lean more, I’ll change my mind, but for now, I’m sticking to my contrarian viewpoint.

• NaiveTortoise 1 Apr 2019 23:33 UTC

  1 point

  in reply to: Rohin Shah’s comment on: What are effec­tive strate­gies for miti­gat­ing the im­pact of acute sleep de­pri­va­tion on cog­ni­tion?

  Good to know I’m not the only one!

  For me, the strategy you describe works sometimes but fails when I block out my schedule for a day to allow myself to focus and then unexpectedly sleep poorly the night before.

• NaiveTortoise 1 Apr 2019 23:34 UTC

  1 point

  in reply to: Sameerishere’s comment on: What are effec­tive strate­gies for miti­gat­ing the im­pact of acute sleep de­pri­va­tion on cog­ni­tion?

  Thanks! These are good recommendations!

  I used to do cold showers but stopped a while back. Maybe I’ll revive the practice.

  Eating low carb and as little as you can

  I’m curious about this one. You find that eating low carb/​less keeps you sharper on days when you sleep very little? I’ve anecdotally noticed that fasting gives me an energy boost on days when I sleep well, although at the cost of making me a bit more jittery, but haven’t noticed the same effects on days when I don’t sleep enough. Is this purely anecdotal or can you share some articles/​papers about this?