18 ms·
Inadequacy and Modesty
- oenat2nteoa0 9y agoWow, that constantly updating page title is irritating. Couldn't even read because of the constant animation at the top of the page.
- adamzerner 9y agoIt's a new site and is on the todo list to fix.
- Afforess 9y agoI really appreciate this essay, if just for categorizing a type of argument I've run into multiple times myself. I've often shared ideas with others and found that when ideas cross a certain "weirdness threshold" and are sufficiently deviant from mainstream thought, the "modest epistemology" argument crops up. After all - if weird idea X was so great, it should already exist. I've never had a proper standard response to this sort objection. Sometimes the key to recognizing something is wrong is the ability to categorize it.
- ianai 9y ago“Knowing is half the battle”
- jancsika 9y agoThat's likely to simply flip your current intuitions in favor of investigating more ideas that pass the weirdness threshold. Without considering the specific context of ideas under discussion, I'm not sure that gets you anywhere.
- neltnerb 9y agoOne convenient metric is "how likely is it that someone with the particular knowledge required to think of this also has the problem I do?"
- bsder 9y agoMaybe. But there are two modalities to this: 1) Implementation often takes one clever person with the requisite knowledge and then everybody copies it if it's a good idea. BitTorrent is a good example. Lots of people could have developed BitTorrent or developed something almost like BitTorrent, but Bram Cohen pulled it all together into one neat, usable package. This modality is probably harder to spot as the "Am I actually clever?" question is difficult to answer with good Bayesian priors. 2) Implementation requires someone with an intersection of skills that limits how many people can exist. My last experience getting an EEG is a good example: the individual pieces of EEG equipment are absolutely primitive to any decent electrical engineer and the user interface to the software is horrifically bad (one step up on reading a paper chart strip). Yet, to improve it, you need an intersection of low-noise analog design, UI programing, signal feature detection and extraction, machine learning applicable to signals, and knowledge of how EEG's work and are used for diagnosis. There simply aren't that many people with all of those characteristics that could move the front forward. However, this modality is VERY easy to spot: "Do I have the intersection of skills necessary to be this person?" is a very straightforward question.
- mcguire 9y ago"Weirdness threshold" is probably not the best term here. Similar arguments crop up often in notably non-weird discussions, such as "if women were really paid less for the same work, a company employing only women would have a major competitive advantage."
- dsacco 9y agoWow. The section beginning from ”If I had to name the single epistemic feat...” and ending at ”...on modern-day Earth.” might be the most riveting description of finance I’ve read in recent memory. If that series of bullet points doesn’t grab someone’s interest and make them want to engage in the robust hyper-competition of modern capitalism, I’m not sure anything will. Really, that entire example of price discovery as a demonstrable core competency and focus of humanity is very interesting to think about. This is a great (and cynical) observation, though it’s much more powerfully delivered in the context of the article: ”Our civilization cares about whether Microsoft stock will be priced at $37.70 or $37.75 tomorrow afternoon.”
- robotresearcher 9y ago> make them want to engage in the robust hyper-competition of modern capitalism But the message of that section is exactly that doing so is irrational, given that enormous resources are already deployed on that problem.
- dsacco 9y agoI guess I should rephrase: my point is that Eliezer made what I would consider one of the most compelling pitches I’ve seen for entering finance if you’re looking for high stakes, intellectually gratifying work.
- tgb 9y agoWell no, the conclusion was that the available inefficiencies should be just about enough to pay the salary of an average person working in that field. Since salaries in that field tend to be really quite good, it's going to far to say that the argument suggests that this is irrational. Instead, it suggests it would be irrational to expect to be able revolutionize the field or otherwise contribute substantially more than those already practicing.
- robotresearcher 9y agoThat's true, and a better statement than mine.
- seibelj 9y agoI have been operating at the intersection of engineering and entrepreneurship for my whole career. Two key traits I notice in entrepreneurs is overwhelming optimism and the ability to take huge risks without batting an eyelash. Conversely, on the engineering side, my fellow nerds are almost paralyzed by over-thinking things and wondering about what the "experts" say, if a decision is "safe", agonizing over career moves, etc. The result is a safe, careful life with above-average to upper-class income, but avoiding the risks that would lead to super-charged results. This essay reads to me like the ultimate nerd analysis of why you should never take any risks. Why start a company when every good idea is taken, and MicroApple AmaGooFace can destroy you the moment you get traction? Why buy any stock when the smartest people on the planet already priced it accurately (surely you don't know more than them)? Why do anything outside of mainstream thought at all? I assume future chapters of his book will address why the rational market hypothesis is incorrect, or why it can be rational to buy AMZN when the market (obviously) knows more than you, but I wanted to comment on this specific post. The truth is that infinite opportunities exist to make money, make an impact, start something worthwhile. However there is a limited supply of people with the balls (or ovaries) to seize these opportunities and take the necessary risks. So many people settle for the cushy life, handcuffed to the big company. The risk takers earn the rewards, and even if they fail, the journey to failure almost always leads to meeting tons of people, tons of opportunities, and a future path that leads back to the cushy high-paying job they tried to avoid. So buy the stock, start the company, assume you know more than the experts, take the risks. You can always go back to the safe life.
- JoshTriplett 9y ago> This essay reads to me like the ultimate nerd analysis of why you should never take any risks. On the contrary, it analyzes in great detail how you can better calibrate your sense for when you should or shouldn't believe you can do better. Which seems like an improvement both over the "you never can" mindset (which you're arguing against and seem to think the article espouses) and the "you always can" mindset.
- tempestn 9y agoThe author is not saying that you shouldn't take any risks or disagree with the masses - nearly the opposite. Essentially the article states that while there are some circumstances where it is true that you almost certainly can't 'beat the market' (the stock market being one), one shouldn't believe that if something was a good idea it would have already been done. Apparently in the next chapter he will lay out a framework for identifying the specific areas that are most likely to present these opportunities.
- unabridged 9y agoThe market price is based on people's intuition multiplied by the money they control. There are plenty of situations where the majority of the money ends up on the opposite side from the smart bettors. Eventually, in a long enough time frame where agents don't die, yes the smart bettors will control most of the money. But the world is imperfect, there are plenty of fools still with money.
- nautilus12 9y agoThis is very eloquent but somewhat verbose and rambling in parts. The only solid take aways i pulled from it are: 1. The assumption that if its easy than any one would have done it in some markets is very true but is not a universal maxim. 2. You have to search out the markets where that is less true. To expand on the station analogy with the $20 bill they mention in the middle of the station where the traffic is highest but what about in the alley behind the station or next to the stand with the new careless employee? Arbitrage isnt just limited to the most trodden popular paths (s&p 500). The context always changes the ebb and flow surrounding arbitrage, whereas you cant fight the raging river and expect to beat it, you may find opportunity in a swirling eddie
- hateduser2 9y agoIt’s not really true. There are real cases where people have radical, obvious insights. Look at history of science and math!
- hyperpallium 9y agoWhat often happens to startups, is they begin with some exciting idea, and having looked into it/tried it, it doesn't pan out, and they "pivot" - sometimes slightly, often wildly. I'm not sure how common this dynamic is in science and maths. Certainly, as you learn more, your ideas and concepts are inevitably refined. (Like: no plan survives contact with the enemy). But I don't know whether that "refinement" is too subtle to count as a "pivot". However, serendipity has a storied past in math and science.
- TeMPOraL 9y agoScience and math are a repeating story of civilization-level of knowledge reaching a threshold at which some breakthroughs become easy to figure out, followed by a bunch of smart people making those breakthroughs simultaneously. I.e. the famous scientists were not that smart - they were just the smartest and therefore first to react to the waterline of knowledge in their domain rising high enough.
- natural219 9y agoEliezer, as always, vastly overstates the value of his own intelligence and intelligence in general. I believe this essay is one of the worst he has ever written. The reasoning on this is correct. The problem is, he is arguing from hilariously inaccurate axioms. For instance, he seems axiomatically convinced that "creating trillions of dollars of value" is an objective good. That is totally fine for people ideologically "all in" to ideologically cooky notions of macroeconomics, which is one of the most intellectually shaky disciplines still in circulation. Since he seems familiar with the "two economists walking down the street" parable, I will offer him this one: === Two economists were walking down the street one day when they passed two large piles of dog turds. The first economist said to the other, "I'll pay you $20,000 to eat one of those piles of turds." The second one agrees and chooses one of the piles and eats it. The first economist pays him his $20,000. Then the second economist says, "I'll pay you $20,000 to eat the other pile of turds." The first one says okay, and eats the turds. The second economist pays him the $20,000. They resume walking down the street. After a while, the second economist says, "You know, I don't feel very good. We both have the same amount of money as when we started. The only difference is we've both eaten turds." The first economist says: "Ah, but you're ignoring the fact that we've engaged in $40,000 worth of trade!"
- Tomminn 9y agoDisagree with your view of the essay, but your anecdote is interesting. I learned something from analyzing the anecdote with inappropriate sincerity. Why- of their own free will- would each economist offer $20,000 to eat the turd? There can essentially only be one answer. They value the schadenfreude or sense of power derived from watching their fellow economist eat the turn at greater than $20,000. Why would each economist eat the turd? They value an extra $20,000 more than they value avoiding the disgust associated with eating the turd. Long story short, that $40,000 of seemingly destructive trade appears to have improved both their lives according to their own value functions.
- natural219 9y agoYou would do well to check if your beliefs are falsifiable. If you believe that GDP objectively, universally, and precisely maps to real-world value, no thought experiment will disabuse you of that notion.
- Tomminn 9y agoMost bad decisions find their origin in either relying to heavily on ones own thoughts, or not sufficiently trusting them. Having a good meta-instinct for knowing when to cede to conventional wisdom and when to trust your own instincts is like a superpower. It's standing on the shoulders of giants and then reaching a little further. Building a strong conceptual foundation around this problem could be really useful.
- donatj 9y agoA truly fascinating read. Actually inspired me to preorder the mentioned book. I’ve never read a book this strongly Econ, but it interests me to no end so this should be an interesting journey.
- skybrian 9y agoI found this article convincing. However, I also found Scott Alexander's article about epistemic learned helplessness convincing: http://squid314.livejournal.com/350090.html http://squid314.livejournal.com/350090.html
- dennisgorelik 9y agoThat Scott Alexander's "Epistemic learned helplessness" blog post (2013) -- is an excellent generic rebuttal against Eliezer's arguments. Eliezer's argumentation is bright, inspiring and mostly right. But then Eliezer makes a critical mistake in his reasoning and the overall conclusion is just plain wrong (such as "buy cryogenics").
- dxdu 9y ago> That Scott Alexander's "Epistemic learned helplessness" blog post (2013) -- is an excellent generic rebuttal against Eliezer's arguments. It's a fully general counterargument against anything.
- kijin 9y agoBeing epistemically helpless, or at least a bit modest, is a good way for an individual to avoid having false beliefs. I don't think this advice is good for the society as a whole, though. Encouraging people to be modest, stand in line, and yield to the majority view by default is a pretty good recipe for a society to become biased toward inertia, often in a very oppressive manner. Japan is one example of such a society. They've made impressive economic and technological progress, but their culture and politics are stuck in a weird loop. The article you linked to also acknowledges that people who are less epistemically helpless are "the only people who can figure out if something basic and unquestionable is wrong, and make this possibility well-known enough that normal people start becoming willing to consider it." Societies need people like this, even though they share some traits with terrorists. So although overconfidence and an experimental spirit might lead individuals to have more false beliefs and failures in the short term, encouraging them might be more beneficial in the long term on a larger scale. This is something that both authors could agree with. I don't know whether Eliezer Yudkowsky would actually be willing to go in this direction, since he looks rather preoccupied with ensuring that each individual human (or AI) makes the best decision at any given time; but since I'm sure he wishes the best for the future of our civilization, I look forward to seeing him pay attention to the social and historical consequences of the individual strategies he's been promoting. A horde of smart robots could behave very differently from a single smart robot if they use the wrong consensus algorithm.
- indubitable 9y agoFor a somewhat contrary view, there is Richard Feynman and Disregard! [1] I tend to fundamentally disagree with the notion of this paper. For a reason why let's consider something like quantum mechanics and relativity. Right now relativity and quantum mechanics are incompatible in domains where they intersect. There is an absolutely enormous incentive to reconcile this. The Nobel Prize would be little more than a formality for a discovery of this magnitude. And the best and brightest in the world have been working on this for many decades. So the implication of this paper is that, thus, you probably shouldn't try to improve the work here. Of course that's nonsensical since it's obviously provably in need of improvement. And there will be some individual, or group of individuals, that manage to succeed here and push humanity that much more forward. He relies on examples that are not falsifiable. The absence of evidence is not evidence of absence. In particular in his stock market example, he argues that it is operating optimally by little more than an appeal to authority and a lack of contrary evidence. It's like saying that no human will be able to run faster than Usain Bolt since lots of people (and massive supporting teams) are spending their lives trying to do just that (with 0 barriers to entry) and all have failed, so there's no reason to believe any given individual could ever do better. Of course the one thing the olympics shows is that records are set only to be broken. I suppose it depends on your life view, but I think people should pursue that which interests them - even if an objective analysis would lead them to believe that their chances of meaningful (and who gets to determine that?) success are negligible. And I would take this a step further. Not only should people do this but the entire progression of humanity is contingent on people doing this. The people that make true progress in a field tend to love that field. And their progress is more a consequence of that love than any sort of higher level objective analysis of fields where their best chances of success lay. Maybe it's best to conclude with a simple counter example. He mentions, repeatedly, you should question yourself if you see a $20 bill laying on the floor in a populated area for hours. And this is supposed to be meaningful. However, instead of relying on a likely very rare outside force (somebody dropping a $20 bill) let's consider something that's an entirely typical and common social phenomena - elevators. How many people have come to a crowded area just outside an elevator, only to find that the problem is that nobody has pushed the button? There's certainly a major incentive to push that button, and there's no barrier to entry other than the arrogance of thinking everybody else might be a bit silly - yet time and again we find nobody pushed the button. [1] - https://stepsandleaps.wordpress.com/2017/10/17/feynmans-breakthrough-disregard-others/ https://stepsandleaps.wordpress.com/2017/10/17/feynmans-brea...
- paulpauper 9y agoWhen we critique a government, we don’t usually get to see what would actually happen if the government took our advice. But in this one case, less than a month after my exchange with John, the Bank of Japan—under the new leadership of Haruhiko Kuroda, and under unprecedented pressure from recently elected Prime Minister Shinzo Abe, who included monetary policy in his campaign platform—embarked on an attempt to print huge amounts of money, with a stated goal of doubling the Japanese money supply.5 Immediately after, Japan experienced real GDP growth of 2.3%, where the previous trend was for falling RGDP. Their economy was operating that far under capacity due to lack of money.6 Given that Japan has been in a slump for nearly 2 decades and that everything tried before didn't work (including stimulus spending), it's reasonable that there would be skepticism.
- mcguire 9y agoThe fundamental assumption of the article seemed to be that the Bank Japan's committee are idiots; any further inquiry into why they make the decision they do stops there.
- abhi3 9y agoDoes this blog have a RSS feed?
- knight17 9y agoYes: https://www.lesserwrong.com/feed.xml https://www.lesserwrong.com/feed.xml. Firefox auto-detected it.
- dennisgorelik 9y agoAll that Eliezer's argumentation is thought provoking and fun. Just keep in mind that Eliezer is good at suggesting mental frameworks for making decisions, but fails when trying to make the best practical decisions himself.
- Nimitz14 9y agoThat was a good read, thanks for posting.
- Bromskloss 9y agoLesserWrong? What is its relation to LessWrong? It talks about itself as "LessWrong 2.0". Is that an official statement or a figure of speech? Edit: This is embarrassing. I had had written "its" as "it's". I need to find a way to do penance.
- purplezooey 9y agoI got a snortchuckle out of "Comments sorted by magical algorithm".
- mbourgon 9y agoNext chapter has been posted: https://equilibriabook.com/an-equilibrium-of-no-free-energy/ https://equilibriabook.com/an-equilibrium-of-no-free-energy/