alkjash

Karma: 4,834

alkjash 26 Nov 2020 19:11 UTC
55 points
0
in reply to: Richard_Ngo’s comment on: Pain is not the unit of Effort
I think the replies to this and the previous post have surprised me in how much even LessWrong readers are capable of rounding off a specific technical claim to the nearest idea they’ve heard before. Let me attempt just once to state what the thesis is again.
I am not saying that effort should never painful. I am also not saying that many useful interventions are painful. I am specifically saying that when you measure effort in units of pain this systematically leads to really bad places (and also that a lot of people are doing this). For example, you will discount forms of effort that are pleasant even if they are more effective.
“If it hurts, you’re probably doing it wrong.” This is just an assertion from an analogy with sports, where even the analogy is false—elite athletes put themselves through a ridiculous amount of pain.
Most amateur and intermediate athletes are doing something wrong if it hurts. “Most amateur and intermediate athletes” is a much larger piece of probability space than “elite athletes.”
As evidence for this, note that the domain where rewards are most heavy-tailed (entrepreneurship) is notorious for being painful: “It’s like chewing glass and staring into the abyss.”
I’m not quite sure I see how this is evidence for your point? Most entrepreneurs fail. It’s possible that they fail because they can’t handle enough pain. It’s also possible that they fail because they Goodharted on a terrible heuristic and lost all the free energy human brains need to innovate. As someone in mathematics research, which I thought would be filled with staring at a wall banging your head until I actually tried it, my gut leans towards the latter.

alkjash 28 Nov 2020 1:47 UTC
48 points
0
on: Pain is not the unit of Effort
I’ve been sitting with this post for a couple days and I’m starting to feel like it is only the tip of the iceberg. Here are three more pieces of the phenomenon to add some color:
1. My brain may be artificially injecting unpleasantness into effort. When engaged in activities that fall into the category of “work” I think my brain adds additional, unnecessary doses of drowsiness, anxiety, and feelings of low status and low agency. While doing the same activity, I can make these feelings disappear by remapping the activity in my head as “play.” I hypothesize that this is an attempt to prove to myself that I am the kind of person who tries hard.
2. In attempts to reintegrate old memories of working hard in school, I feel a mental flinch every time I suggest the hypothesis that “In this instance, I put myself through a ton of pain for no reason.” I predict that the primary immune reaction people will have to reading this post is the feeling of “This can’t be true, or else all my suffering would have been pointless!” People like very much to ascribe meaning to suffering. This also maps onto behaviors like “Back in my day...” complaining and slapping people down for looking for easy weight loss solutions “because losing weight is supposed to be hard.”
3. Part of the “pain is the unit of effort” heuristic is that pain is supposed to be the signal that you’ve exhausted your mana bar. Part of why it’s so attractive is the idea that if you don’t spend your mana, you’re wasting it. So you work up to your pain tolerance to spend it all. My rebuttal to this model is that human energy is really frigging weird and a fixed mana bar is not a good model for it. I can have an entire afternoon of small group meetings and come out of them feeling more energized than I went in. Others have claimed that human energy is even weirder; for example in the discussions around Kensho I recall Valentine making a claim (can’t find the exact place) indistinguishable from “There is a way to input ↑↑↓↓←→←→BA into the brain through meditation practices that unlocks infinite energy mode.”

alkjash 25 Nov 2020 18:20 UTC
43 points
2
in reply to: digital_carver’s comment on: Pain is not the unit of Effort
I think this is an important consideration, and the picture does look different if you are living with chronic (physical or psychic) pain, but I disagree with the overall claim.
When your very survival, barely holding together a family and a home, takes up 15 hours of your day, it’s going to hurt if you’re trying to advance yourself with education or other self-improvement, it’s going to take a necessary toll on your physical and mental health.
This seems like an over-generalization, and does not mesh with my experience talking to family members who escaped poverty. If anything they found escape and respite in working on their education, and having hope and a direction to work towards increased their subjective well-being. The mindset may be determined by personality differences more than differences in circumstance.
I think what is true is that when you are carrying a huge amount of pain and urgency in your body day-to-day, you become numb to the amount of annoyance that is “studying for the fourth hour in a row” or “sharp pain in the wrists from bench pressing.” Depending on the situation this can be either marginally effective or actively dangerous, but regardless pain is not an accurate unit of effort or progress here. Perhaps the fourth hour of studying causes more than half of the total pain, but it is highly unlikely that the fourth hour of studying was responsible for more than half of the total learning that occurred. Of course, the fourth hour could very well still be worth doing, simply because any marginal speedup on “getting out of my current situation” is worth massive amounts of temporary unpleasantness.

alkjash 23 Feb 2018 21:55 UTC
36 points
on: Two types of mathematician
There’s an obvious but cool meta-thing going on with the number 2 that might be useful to pick out. Some pieces of this thing:
All over the place, we speak in terms of dichotomies and not trichotomies or more. The reason is basically that each dichotomy corresponds to doing PCA and projecting space onto a single axis, and a one-dimensional line has two directions. This suggests that much of the interesting conversation about any given topic (i.e. axis) can be picked up by having exactly two people talk about it. Two people will always differ slightly on the axis. Adding any additional people to a conversation has rapidly diminishing returns: you may have more total disagreement, but rarely more total dimensionality in the disagreement.
Duo Talks is the idea that most productive conversations occur between two people. Even with Trio walks, the setup is for one person to stay on the sidelines and wait for a chance to rotate in. A single conversation really only happens along one dimension, and it requires only two distinct people to detect.
I wonder if monogamy is some kind of attractor state because interactions between two people are the most productive.
Applying the Solitaire Principle, for all the same reasons it’s useful to have conversations between two people, it’s most useful to draw dichotomies between two pieces of the mind instead of more. This is why we have Internal Double Crux instead of Triple or more. A conversation/internal conflict is always about something and should have a purpose, and that purpose projects the entire conversation onto the one relevant dimension, so it’s really only necessary to divide into two sides along this axis. Thus we get S1/S2, Elephant/Rider, Episodic/Diachronic, etc.

alkjash 24 Feb 2018 1:00 UTC
27 points
on: Mythic Mode
Here’s a seed of a possible future post, in response to the introductory part about Moloch:
The Four Demon Gods and the Psychologists Out to Stop Them
The bowels of Hell are divided by two rivers of flame into four quadrants.
The left-right axis reads Weakness on the left and Strength on the right. Weakness demons prey on cowardice and jealousy. Strength demons dominate and exploit.
The up-down axis reads Private on the bottom and Public on top. Private demons live in the hearts of individual human beings. Public demons live in the empty voids between people.
Our tour of Hell takes us clockwise around this wheel of Evil from the top left quadrant.
(1) The top left quadrant is the domain of Moloch. Moloch is Public Demon of Weakness. The decay of whole cities down incentive slopes, unbeknownst to their citizens. The whisper of defect, defect on frigid prison nights. The tragic coordination failures of naive libertarians. God’s chosen champion against Moloch is Scott Alexander.
(2) The top right quadrant is the domain of The Man, known sometimes as the Patriarchy and Big Brother. The Man is the Public Demon of Strength. The Man is Out to Get You for all you’ve got. The Man has his grubby fingers around the throat of every public conversation, framing the Overton Window however he wishes. God’s chosen champion against The Man is The Last Psychiatrist.
(3) The bottom right quadrant is the domain of Satan. Satan is the Private Demon of Strength. He is the perfectionist critic in your soul, the fallen angel of Justice. He judges the nature of being itself unworthy and screams for its destruction. God’s chosen champion against Satan is Jordan Peterson.
(4) The bottom left quadrant is the domain of The Underground Man. The Underground Man is the Private Demon of Weakness. He is a sick man, a spiteful man. He is the decay of the human soul into the basement-dweller and the Nice Guy. He is the weakness in your soul that allows things to crumble. God’s chosen champion against The Underground Man is Fyodor Dostoevsky.

alkjash 16 Dec 2020 5:05 UTC
25 points
on: Motive Ambiguity
I wonder if the following are also examples of motive ambiguity:
- Mothers choosing to stay at home.
- Researchers choosing to be bad at teaching.
- Mathematicians choosing to work on problems with no applications.

alkjash 26 Nov 2020 0:00 UTC
25 points
on: Pain is the unit of Effort
They say imitation is the highest praise so thanks for writing this. :D
A substantial chunk of my heart identifies with this post, but what I think it’s pointing to is that pain can be a reasonable measure for a certain kind of effort. The difference between this post and mine feels close to the difference between making a desperate effort and making an extraordinary effort.
When I imagine making a desperate effort, it feels like exactly what your post describes: pushing my body to the point of physical exhaustion and reaching a level of achievement I didn’t think possible. When I imagine making an extraordinary effort, I see instead: sitting back in an armchair and surreptitiously saving the world by writing a fanfiction while following my heart’s desire. Reality doesn’t care how much pain you’re in. Perhaps you can tell by my language, but the latter is what I aspire to.
If you live in a world where nobody is trying and you find in yourself the ability to try in one particular way by making a desperate effort, your body might learn that this is the only way to try. I would offer that if you sat back and did only things that came easily, you might learn an altogether different way of trying with all the extra slack that saves you.

alkjash 8 Feb 2018 18:41 UTC
25 points
on: Write a Thousand Roads to Rome
I had a good laugh reading the negative reviews of the Sequences on Goodreads. For every ten silent aspiring rationalists loving it exactly the way it is, there’s one vocal critic saying things like:
The book is very informative and I really liked some parts, but it was a very difficult read. Why? Because Yudkowsky simply LOVES beating a dead horse. There’s just too much repetition and sloppy writing, although the ideas are great.
It’s just… It ought to be more concise and clear.
This one was really hard to rate as my reactions to each post and sequence varied from “jeez, this just nailed it” to “meh, why are you explaining the same thing over and over again”.
Eliezer knew exactly how important it is to explain the same thing using different words again and again.

alkjash 29 Oct 2020 0:46 UTC
24 points
in reply to: Viliam’s comment on: Is Success the Enemy of Freedom? (Full)
Wow, this comment alone makes writing the entire post worth it.

alkjash 21 Mar 2018 23:20 UTC
23 points
on: The Math Learning Experiment
Cool experiment, I hope you run more things like this!
Re: definitions, this is a serious struggle I have with the way math is presented. The definitions and notations we have now have been simplified, condensed, and iterated upon for hundreds of years, and what sticks is a matter of practical utility for mathematicians. To present the product of this process as a fait accompli without explanation or motivation is really misleading.
I mean, look how long it took to come up with the modern definition of the group! Even after it was named, for the longest time, mathematicians only considered permutation groups.
I would lay some of the blame at the feet of Platonism. Mathematical Platonism, the idea that “mathematics is out there in idea-space and we just discover it” is at best a useful metaphor or fake framework and has been taken way too far. It implies that definitions are somehow floating in thought-space and the mental move to discover them is to look really hard. In practice, the construction of definitions is more like engineering: there’s some weird thing that’s there and we’ll try to build a vessel that captures its shape as closely as possible.
Note: I tried verifying the received wisdom that “most mathematicians are Platonists” and found the answer to be much murkier than I expected. I would still say that mathematics, the way it is presented, suggests that things like chain complexes and schemes are “things out there in reality” instead of “definitions progressively engineered by people to fit reality.”

alkjash 3 Dec 2020 2:37 UTC
21 points
on: alkjash’s Shortform
Instrumental Rationality Mini-Retrospective
I promised several years ago to write a retrospective on Hammertime a year after it was released. I broke that promise but I wanted to take some time to do the work now, and to summarize my current beliefs about how much rationalist self-improvement affected my personal growth. I’d also like to estimate how it compares to other schools of self-improvement I’ve dabbled in.
First, I should mention that epistemic rationality has been directly useful in my career, although this is highly unlikely to generalize. At least three of the best papers I wrote in graduate school featured tricky applications of Bayes’ Theorem. I was able to do from muscle memory certain calculations about conditional probability and expectation that might have taken weeks otherwise (if we figured them out at all). I attribute this ability in large part to reading the Sequences. YMMV if you don’t work in or adjacent to probability theory. The rest of this review will ignore this relatively idiosyncratic improvement and focus on my beliefs about what is usually categorized as “instrumental rationality.”
Over the last three years, I’ve seriously attempted the following “self-help schools of thought”: CFAR applied rationality, therapy, Jordan Peterson, Tony Robbins. Insofar as numbers mean anything, in terms of personal progress in my life: CFAR is responsible for a ~100% improvement, therapy ~15%, Jordan Peterson ~40%, and Tony Robbins ~200%. Tony Robbins I encountered quite recently so the error bars are the biggest, and the effects have not yet had time to play out. As weak but verifiable evidence that the other three actually made an impact on my life, my publishing rate at least doubled since CFAR, the theorems have improved despite this. Also, I believe the quality of my writing has radically improved in these three years.
For a few more gears into these numbers:
CFAR is unique in that it (a) synthesizes the best bits of many other schools and couches it in language that is palatable to its target audience, (b) immerses you in practice in a way that any other immersive workshop does approximately equally well, but is nevertheless strong in a way regular practice and habit isn’t, and (c) is astronomically memetically safe compared to the alternatives, due to deliberate choices like noting when models are “fake but useful but definitely fake.”
Therapy can be hit-or-miss, the average-case scenario seems to be roughly that you pay a savvy friend armed with one or two CFAR techniques to listen and consciously apply that technique once a week. Still probably worth it.
Jordan Peterson was very useful as a charismatic psychology professor. His Youtube lectures seem to be the best available resources on the subject, and I don’t believe there is a single other source that will get you the same level of understanding of the academic literature at the same cost in effort. What you get out of that abstract understanding will depend primarily on your general problem-solving ability. I don’t believe there’s anything remarkable about his self-improvement philosophy, although doing it will help (just because doing anything like this helps).
Tony Robbins is the best-in-class person at simultaneously blasting every part of your brain with charisma and motivation, and works for a general audience of homo sapiens (unlike CFAR). He is also extremely memetically dangerous—he knows all the buttons of the human brain that have been discovered and will push whichever ones it takes to effect the change he envisions. (For example, much of his technique rests on attaching positive meaning to suffering.) If it’s any comfort, his intentions seem to be about as good as one can possibly expect for someone armed from adolescence with this superweapon. He is currently my leading candidate for “human best at getting out of the AI box.” I do not believe I could have safely jumped into watching a bunch of his videos, learned a lot, and jumped back out, without much of the preparation and growth I got from the other sources listed above.

alkjash 3 Dec 2020 19:04 UTC
20 points
in reply to: DirectedEvolution’s comment on: Pain is not the unit of Effort
Your comment is interesting and helpful for me because I have only a small sample of people who don’t follow the “pain is the unit of effort” heuristic to a pathological extent. Perhaps this is explained by my circle of friends being dominated by Asian-Americans who went to the top universities. I definitely didn’t consider what possibly ill effects it might have on others for whom this is not true. So thanks for that information!
However, enough of my brain interpreted your comment as a status move/slapdown that I’d suggest you reconsider doing these reviews, at least in the tone you’re currently doing them. I don’t believe you intended it this way, but your comment comes off as claiming a position of authority and also encourages too much (imo) Goodharting on the LessWrong “top 15 posts” metric. Both of these feel icky to me. I predict you will at minimum annoy a lot of authors if you continue to write these.

alkjash 11 Mar 2018 20:05 UTC
20 points
in reply to: cousin_it’s comment on: Misery Pits
Several nasty things happened in one week: we split up with my first gf after a 7 year relationship (mostly due to my neediness), the company I worked at went bankrupt, I lost my apartment, and my grandmother died in the hospice as I watched. That wouldn’t bother a stable person much, but 23 year old me also wasn’t a very stable person.
Jesus, whoever these “stable people” you speak of are, I’d like to meet them. I don’t think I know a single person who wouldn’t crack in such a situation.

alkjash 10 Sep 2019 20:46 UTC
19 points
on: [Hammertime Final Exam] Quantum Walk, Oracles and Sunk Meaning
I just stumbled back into LessWrong after many months, very pleased to see that you completed this journey with such seriousness! Looking forward to your thoughts and comments in the postmortem post.
Re: Sunk Meaning, immediately after reading this, my brain sent strong signals that *something else is going on here* with regards to making the cow’s death mean something and there’s some Chesterton’s fence that you shouldn’t break. I think the error you’re making is basically the same error that causal decision theorists make two-boxing in Newcomb’s problem. You do not get to choose whether or not to honor a dead cow in this specific instance. You only get to choose (for all time) whether or not your source code implements an algorithm which honors the dead. Being able to do this is really really useful for all sorts of Newcomblike situations that occur in real life.

alkjash 6 Feb 2018 22:07 UTC
19 points
on: “Cheat to Win”: Engineering Positive Social Feedback
I think this is very good advice, but I’d like to push back on the “haters don’t matter” bit. One consistently negative person can do a lot of damage, especially if they punish you for your virtues. I’d probably pay even more attention to removing people from your social circle who punish you for the things you want to do than adding people who reward you for them. Certain classes of haters are also nearly impossible to avoid.

alkjash 23 Feb 2018 20:38 UTC
17 points
on: Two types of mathematician
Very pleased to see all of these dichotomies collected in one place. The natural question is whether these divides can be integrated to a useful picture with more pieces.
My take on the “Two Cultures” model of problem-solvers and theory-builders: theory-building fields of mathematics like algebraic topology (say) are those where the goal is to articulate grand meta-theorems that are bigger than any particular application. This was the work of a Grothendieck.
Meanwhile, concrete problem-solving fields of mathematics like combinatorics are those where the goal is to become the grand meta-theorem that contains more understanding than any particular theorem you can prove. This was the style of an Erdos. The inarticulate grand meta-theorems lived in his cognitive strategies so that the theorems he actually proved are individually only faint impressions thereof.

alkjash 13 Mar 2021 3:36 UTC
15 points
on: Trapped Priors As A Basic Problem Of Rationality
Is the following interpretation equivalent to the point?

It can be systematically incorrect to “update on evidence.” What my brain experiences as “evidence” is actually “an approximation of the posterior.” Thus, the actual dog is [1% scary], but my prior says dogs are [99% scary], I experience the dog as [98% scary] which my brain rounds back to [99% scary]. And so I get more evidence that I am right.

alkjash 24 Feb 2018 22:11 UTC
15 points
in reply to: cousin_it’s comment on: Mythic Mode
I’m bad and I feel bad about making this kind of argument:
I don’t want powerful weapons to argue for truth. I want asymmetric weapons that only the truth can use. Myth isn’t such a weapon, so I’ll leave it in the cave where it was found.
Register the irony of framing your refusal to use the power of mythical language in a metaphor about a wise and humble hero leaving Excalibur in the cave where it was found.
The issue is that we are all being pulled by Omega’s web into roles, and the choice is not whether or not to partake in some role, but whether or not to use the role we inhabit to our advantage. You don’t get to choose not to play the game, but you do get to pick your position.

alkjash 31 Jan 2018 22:30 UTC
15 points
in reply to: Swerve’s comment on: Hammertime Day 1: Bug Hunt
I wouldn’t worry about it, 50 is plenty for now. My hope is that the first cycle will cultivate stronger lateral thinking and expand your comfort zone enough that you’ll be able to come up with plenty more on the next cycle.
There’s a larger point I want to make which is that the reason one wants to be a rationalist is to win in general. If there were a particular one or two domains you care about, then applying force on those domains will get you more for your dollar than working on rationality as such (unless the particular domain has something to do with minds). Therefore, to make maximal use out of rationality you have to be solving lots of disparate problems, which requires that you notice lots of bugs to solve in the first place.
Seconding Qiaochu’s point to an extent: huge families of bugs can all be manifestations of one “real” bug at the bottom. On the other hand I have the sense that the best way to get started on these “real” bugs is to solve some of their small consequences, so being able to fractionate bugs into specific pieces is still useful.

alkjash 6 Apr 2021 21:33 UTC
14 points
in reply to: Emiya’s comment on: Internal Double Crux
This is fascinating and I’d love to hear more depth on whatever you’d be willing to share.
Regarding the suggestion to start with something small, I think in hindsight it was kind of a manipulation on my part to make the tool seem safer and to try to get more people to try it. In my limited experience, internal conflicts that seem small rarely turn out to be.
When I first tried IDC at CFAR, the initial “small starting point” of “Should I floss?” dredged up a whole complex about distrust of doctors in particular and authority in general. A typical experience with watching myself and others IDC is that regardless of the starting point, one ends up in a grand dramatic battle of angels and demons over one’s soul.