Posts
Comments
I agree. My reason for posting the link here is as reality check-- LW seems to be full of people firmly convinced that brain-uploading is the only only viable path to preserving consciousness, as if the implementation "details" were an almost-solved problem.
Finally, someone with a clue about biology tells it like it is about brain uploading
http://mathbabe.org/2015/10/20/guest-post-dirty-rant-about-the-human-brain-project/
In reading this, suggest being on guard against own impulse to find excuses to dismiss the arguments presented because they call into question some beliefs that seem to be deeply held by many in this community.
It depends. Writing a paper is not a realtime activity. Answering a free-response question can be. Proving a complex theorem is not a realtime activity, solving a basic math problem can be. It's a matter of calibrating the question difficulty so that is can be answered within the (soft) time-limits of an interview. Part of that calibration is letting the applicant "choose their weapon". Another part of it is letting them use the internet to look up anything they need to.
Our lead dev has passed this test, as has my summer grad student. There are two applicants being called back for second interviews (but the position is still open and it is not too late) who passed during their first interviews. Just to make sure, I first gave it to my 14 year old son and he nailed it in under half an hour.
Correct, this is a staff programmer posting. Not faculty or post-doc (though when/if we do open a post-doc position, we'll be doing coding tests for that also, due to recent experiences).
Having a track-record of contributions github/bitbucket/sourceforge/rforge would be a very strong qualification. However, few applicants have this. It's a less stringent requirement that they at least show that they can... you know... program.
it's not strictly an AI problem-- any sufficiently rapid optimization process bears the risk of irretrievably converging on an optimum nobody likes before anybody can intervene with an updated optimization target.
individual and property rights are not rigorously specified enough to be a sufficient safeguard against bad outcomes even in an economy moving at human speeds
in other words the science of getting what we ask for advances faster than the science of figuring out what to ask for
(Note that transforming a sufficiently well specified statistical model into a lossless data compressor is a solved problem, and the solution is called arithmetic encoding - I can give you my implementation, or you can find one on the web.
The unsolved problems are the ones hiding behind the token "sufficiently well specified statistical model".
That said, thanks for the pointer to arithmetic encoding, that may be useful in the future.
The point isn't understanding Bayes theorem. The point is methods that use Bayes theorem. My own statistics prof said that a lot of medical people don't use Bayes because it usually leads to more complicated math.
To me, the biggest problem with Bayes theorem or any other fundamental statistical concept, frequentist or not, is adapting it to specific, complex, real-life problems and finding ways to test its validity under real-world constraints. This tends to require a thorough understanding of both statistics and the problem domain.
That's not the skill that's taught in a statistics degree.
Not explicitly, no. My only evidence is anecdotal. The statisticians and programmers I've talked to appear to overall be more rigorous in their thinking than biologists. Or at least better able to rigorously articulate their ideas (the Achilles heel of statisticians and programmers is that they systematically underestimate the complexity of biological systems, but that's a different topic). I found that my own thinking became more organized and thorough over the course of my statistical training.
Also, I'm not sure if this is your intention, but it seems to me that the goal of spending 20 years to slow or prevent aging is a recipe for wasting time. It's such an ambitious goal that so many people are already working on, any one researcher is unlikely to put a measurable dent in it.
In the last five years the NIH (National Institutes of Health) has never spent more than 2% of its budget on aging research. To a first approximation, the availability of grant support is proportional to the number of academic researchers, or at least to the amount of academic research effort being put into a problem. This is evidence against aging already getting enough attention. Especially considering that age is a major risk factor for just about every disease. It's as if we tried to treat AIDS by spending 2% on HIV research and 98% on all the hundreds of opportunistic infections that are the proximal causes any individual AIDS patient's death. I would think that curing several hundred proximal problems is more ambitious than trying to understand and intervene in a few underlying causes.
I have no illusions of single-handedly curing aging in the next two decades. I will be as satisfied as any other stiff in the cryofacility if I manage remove one or more major road-blocks to a practical anti-aging intervention or at least a well-defined and valid mechanistic model of aging.
Secondly, you probably shouldn't worry about pursuing a project in which your already-collected data is useless, especially if that data or similar is also available to most other researchers in your field (if not, it would be very useful for you to try to make that data available to others who could do something with it). You're probably more likely to make progress with interesting new data than interesting old data.
This is 'new' data in the sense that it is only now becoming available for research purposes, and if I have my way, it is going to be in a very flexible and analysis-friendly format. It is the core mission of my team to make the data available to researchers (insofar as permitted by law, patients' right to privacy, and contractual obligations to the owners of the data).
If I ran "academia", tool and method development would take at least as much priority as traditional hypothesis-driven research. I think a major take-home message of LW is that hypotheses are a dime a dozen-- what we need are practical ways to rank them and update their rankings on new data. A good tool that lets you crank through thousands of hypotheses is worth a lot more than any individual hypothesis. I have all kinds of fun ideas for tools.
But for the purposes of this post, I'm assuming that I'm stuck with the academia we have, I have access to a large anonymized clinical dataset, and I want to make the best possible use of it (I'll address your points about aging as a choice of topic in a separate reply).
The academia we're stuck with (at least in the biomedical field) effectively requires faculty to have a research plan describable by "Determine whether FOO is true or false" rather than "Create a FOO that does BAR".
So the nobrainer approach would be for me to take the tool I most want to develop, slap some age-related disease onto it as a motivating use-case, and make that my grant. But, this optimizes for the wrong thing-- I don't want to find excuses for engaging in fascinating intellectual exercises. I want to find the problems with the greatest potential to advance human longevity, and then bring my assets to bear on those problems even if the work turns out to be uglier and more tedious than my ideal informatics project.
The reason I'm asking for the LW community's perspective on what's on the critical path to human longevity is that I spent too much time around excuse-driven^H^H^H hypothesis-driven research to put too much faith in my own intuitions about what problems need to be solved.
Great idea! Here's how I can convert your prospective experiment into retrospective ones:
Comparing hazard functions for individuals with diagnoses of infertility versus individuals who originally enter the clinic record system due to a routine checkup.
Thanks for reminding me about SENS and de Grey, I should email him. I should reach out to all the smart people in the research community I know well enough to randomly pester and collect their opinions on this.
People gain skills by working on hard problems, so it doesn't seem necessary for you to take additional time to explicitly hone your skill set before starting on any project(s) that you want to work on.
The embarrassing truth is I spent so much time cramming stuff into my brain while trying to survive in academia that until now I haven't really had time to think about the big picture. I just vectored toward what at any given point seemed like the direction that would give me the most options for tackling the aging problem. Now I'm finally as close to an optimal starting point as I can reasonably expect and the time has come to confront the question: "now what"?
So, for a retrospective approach with existing data, I could try to find a constellation of proxy variables in the ICD9 V-codes and maybe some lab values suggestive of basically healthy patients who consume a lower-than-typical amount of calories. Not in a very health-conscious part of the country though, so unlikely that a large number of patients would do this on purpose, let alone one specific fasting strategy.
Now, something I could do is team up with a local dietician or endocrinologist and recruit patients to try calorie restriction.
I should clarify something: the types of problems I can most efficiently tackle are retrospective analysis of already-collected data.
Prospective clinical and animal studies are not out of the question, but given the investment in infrastructure and regulatory compliance they would need, these would have to be collaborations with researchers already pursuing such studies. This is on the table, but does not leverage the clinical data I already have (unless, in the case of clinical researchers, they are already at my institution or an affiliated one).
My idea at the moment is to fit a hidden Markov model and derive a state model for human aging. But this pile of clinical data I have has got to be useful for all kinds of other aging-related questions...
If we are in an environment of open conversation and I say something that brings up an emotional trauma in another person and that person doesn't have the self-awareness to know why he's feeling unwell, that's not a good time to leave him alone.
?! Depends. If you don't understand that person intimately or aren't experienced at helping less self-aware (aka neurotypical) people process emotional trauma, it's probably a very good time to leave him alone. Politely.
I was tempted to vote "makes no sense at all". I did not because I've had far too many experiences where I dismiss a colleague's idea as being the product of muddled thinking only to later realize that a) the idea makes sense, they just didn't know how to express it clearly or b) the idea makes practical sense but my profession chooses to sweep it under the rug because it's too inconvenient. On Stackoverflow and LW I see the same tendency to mistake hard/tedious problems for meaningless problems and "solve" the problem by prematurely claiming to have dissolved the question or substituting in a different question the respondent finds more convenient.
Some questions really are meaningless or misguided. But experience has taught me to usually give questions the benefit of a doubt until I have enough background information to be more sure. So, I played along and gave the technically correct answer of "I'm parts both".
Come to think of it, "Red/Blue makes no sense at all" is not even a valid answer to the question. The question did not ask whether it made sense. Such a meta-question should really be a checkbox orthogonal to the main poll question.
This has taught me that I find it more intuitive to think in terms of conditional probabilities than marginal probabilities.
The tough part will be guarding against Goodhart's Law. I suspect that the current system of publications and grant money as an indicator of ability started out as an attempt to improve the efficiency of scientific progress and has by now been thoroughly Goodharted.
As Lumifer points out, tenure was intended to give productive scientists some protected time so they could think. However, the amount of hoops you jump through on the way to getting there puts you through the opposite of protected time so by the time you get tenure you've gotten jaded, cynical, and acquired some habits useful for academic survival but harmful to academic excellence.
I can offer advice on statistical analysis of data (frequentist, alas, still learning Bayesian methods myself so not ready to advise on that). Unfortunately, right now I have too little spare time to actually analyze it for you, but I can explain to you how you can tackle it using open source tools and try to point you toward further reading focused on the specific problem you're trying to solve. In the medium-future I hope to have my online data analysis app stable enough to post here, but this is not looking like the month when it will happen.
I can probably answer almost any question you have about the R language, many questions about the Shiny framework, and some questions about Javascript, PHP, and various flavors of SQL (though there are probably plenty LW-ers more knowledgeable than I on the latter three topics).
Also can advise on designing controlled animal experiments so that you won't regret painting yourself into a corner later, but I'm guessing there aren't many biologists here.
I apologize in advance for slow turnaround times. My schedule is pretty full of kids and work. :-/
PS: if your question is too lengthy to post here, just post it on the appropriate Stackexchange site and post the link here.
The current 500-year window needs to be be VERY typical if it's the main evidence in support of the statement that "even with no singularity technological advance is a normal part of our society".
This is like someone in the 1990s saying that constantly increasing share price "is a normal part of Microsoft".
I think technological progress is desirable and hope that it will continue for a long time. All I'm saying is that being overconfident about future rates of technological progress is one of this community's most glaring weaknesses.
Take any 500-year window that contains the year 2014. How typical would you say it is of all 500-year intervals during which tool-using humans existed?
even with no singularity technological advance is a normal part of our society
Depends what time scale you're talking about.
It would look like a failure to adequately discount for inferential chain length.
having people around who give a damn about you
Yes, exactly. I'd add:
...because the best cryopreservation arrangements won't do you much good if nobody notices you died until the neighbors complain about the smell.
Somewhere between an extended family and a kindergarten; like a small private kindergarten where the parents are close friends with the caretakers.
That, right there, is one of my fondest dreams. To get my tiny scientists out of the conformity-factory and someplace where they can flourish (even more). Man, if this was happening in my town, in a heartbeat I'd rearrange my work schedule to spend part of the week being a homeschooler.
dealing with resource scarcity -- and you keep on bringing up how markets don't solve violence and pollution...
Well, they should provide a constructive alternative to the former, and the latter is isomorphous with a scarcity of non-polluted air/water/land.
Here's what I expect someone who seriously believed that markets will handle it would sound like:
"Wow, overpopulation is a threat? Clearly there are inefficiencies the rest of the market is too stupid to exploit. Let's see if I can get rich by figuring out where these inefficiencies are and how to exploit them."
Whereas "the markets will handle it, period, full stop" is not a belief, it's an excuse.
it's only a question of how many planets we consume before that happens.
Hopefully more than one. There are a lot of underutilized planets out there, even within our own solar system.
Fixed, thanks.
The choke point in our Fritz Haber/Norman Borlaug/Edward Jenner pipeline is not the amount of science education out there. It's a combination of the low-hanging fruit being picked, insufficient investment in novel approaches and not enough geniuses.
Very true. Each year we produce thousands of new Ph.D.s and import thousands more, while slowly choking off funding for basic research, so they languish in a post-doc holding pattern until many of them give up and go do something less innovative but safer.
Alternatively tutoring is free and with a similar level of time costs to raising your own children you could tutor a lot of others.
Yes! The school system in my state spends far more on remedial education than on GT. Education is seen as a status symbol instead of a costly investment that should be allocated in a manner that gives the highest returns (in terms of innovation, prosperity, and sane policy decisions).
All of these "what you should do if you are a utilitarian" articles should start with "Assuming you are a being for whom utility matters roughly equally regardless of who experiences it..."
Yes! Thank you for articulating in one sentence what I haven't been able to in a dozen posts.
You should repeat this at the top level. This changes things quite a bit.
We should be careful to make the distinction between jkaufman's own opinions and those of the paper they posted a link to.
By the way, it's refreshing to see people be honest with themselves and others about what they value instead of the posturing/kool-aid one often sees around this topic.
Oops, you're right. I have now revised it.
A witty quote from an great book by a brilliant author is awesome, but does not have the status of any sort of law.
What do we mean by "normality"? What you observe around you every day? If you are wrong about the unobserved causal mechanisms underlying your observations, you will make wrong decisions. If you walk on hot coals because you believe God will not let you burn, the normality that quantum mechanics adds up to diverges enough from your normality that there will be tangible consequences. Are goals part of normality? If not, they certainly depend on assumptions you make about your model of normality. Either way, when you discover that God can't/won't make you fireproof, some subset of your goals will (and should) come tumbling down. This too has tangible consequences.
Some subset of the remaining goals relies on more subtle errors in your model of normality and they too will at some point crumble.
What evidence do we have that any goals at all are stable at every level? Why should the goals of a massive blob of atoms have such a universality?
I can see the point of "it all adds up to normality" if you're encouraging someone to not be reluctant to learn new facts. But how does it help answer the question of "what goal do we pursue if we find proof that all our goals are bullshit"?
So, looking at shminux' post above, you would suggest mandatory insemination of only some fertile females and reducing subsistence to slightly above the minimum acceptable caloric levels..?
I believe that deliberately increasing population growth is specifically the opposite direction of the one we should be aiming toward if we are to maximize any utility function that penalizes die-offs, at least as long as we are strictly confined to one planet. I was just more interested in the more general point shminux raised about repugnant conclusions and wanted to address that instead of the specifics of this particular repugnant conclusion.
I think the way to maximize the "human integral" is to find the rate of population change at which our chances surviving long enough and ramping up our technological capabilities fast enough to colonize the solar system. That, in turn, will be bounded from above by population growth rates that risk overshoot, civilizational collapse, and die-off and bounded from below by the critical mass necessary for optimum technological progress and the minimum viable population. My guess is that the first of these is the more proximal one.
At any rate, we have to have some better-than-nothing way of handling repugnant conclusions that doesn't amount to doing nothing and waiting for someone else to come up with all the answers. I also think it's important to distinguish between optima that are inherently repugnant versus optima that can be non-repugnant but we haven't been able to think of a non-repugnant path to get from here to there.
I don't understand the response. Are you saying that the reason you don't have an egocentric world view and I do is in some way because of kin selection?
How about this as a rule of thumb, pending something more formal:
If a particular reallocation of resources/priority/etc. seems optimal, look for a point in the solution space between there and the status quo that is more optimal than the status quo, go for that point, and re-evaluate from there.
You might be right. I hope not, though, because that means it will take even longer to escape from the planetary cycle of overshoot and collapse.
Then again, it's good to be ready for the worst and be pleasantly surprised if things turn out better than expected.
Once we've dealt with the mass starvation, vast numbers of deaths from malaria, horrendous poverty, etc., then we can start paying a lot more attention to awesomeness.
What if, for practical purposes, there is an inexhaustible supply of suck? What if we can't deal with it once and for all and then turn our attention to the fun stuff?
So, judging from the reception of my post about the Malthusian Crunch certain Wrongians sense this and have gone into denial (perhaps, if they're honest with themselves, privately admitting the hope that if they ignore the starving masses long enough, they will go away).
I propose a middle ground between giving everything and giving nothing-- a non-arbitrary cutoff for how much help is enough. A cutoff that can be defended on pragmatic grounds without having to assume a shared normative morality.
You put just enough resources into pure suckiness remediation to insure that spillover suckiness will not derail your awesomeness plans. I emphasize pure because there are pursuits that simultaneously strive for new heights of awesomeness and fix suck in equal measure. Obviously this quality is desirable and such projects should not be penalized for having it.
But, in any case, I would expect this to lead to the Malthusian scenario we should be trying to avoid, not an overall maximization of all humans who have ever lived.
What if the reason repugnant conclusions come up is that we only have estimates of our real utility functions which are an adequate fit over most of the parameter space but diverge from true utility under certain boundary conditions?
If so, either don't feel shame about having to patch your utility function with one that does fit better in that region of the parameter space... or aim for a non-maximal but closer to maximum utility that that is far enough from the boundary that it can be relied on.
So your issue is that a copy of you is not you? And you would treat star trek-like transporter beams as murder?
Nothing so melodramatic, but I wouldn't use them. UNLESS they were in fact manipulating my wave function directly somehow causing my amplitude to increase in one place and decrease in another. Probably not what the screenplay writers had in mind, though.
But you are OK with a gradual replacement of your brain, just not with a complete one?
Maybe even a complete one eventually. If the vast majority of my cognition has migrated to the synthetic regions, it may not seem as much of a loss when parts of the biological brain break down and have to be replaced. Hard to speak on behalf of my future self with only what I know now. This is speculation.
How fast would the parts need to be replaced to preserve this "experience of continuity"?
This is an empirical question that could be answered if/when it becomes possible perform for real the thought experiment I described (the second one, with the blank brain being attached to the existing brain).
Basically, what I am unclear on is whether your issue is continuity of experience or cloning.
Continuity. I'm not opposed to non-destructive copies of me, but I don't see them as inherently beneficial to me either.
http://lesswrong.com/lw/iul/looking_for_opinions_of_people_like_nick_bostrom/9x47
That should be a new discussion.
You claimed that people ignore or outright oppose trying to accelerate the rate of technological advancement. Could it be instead that nobody has any idea how to do it?
Very, very possible.
An independent settlement seems quite beyond the possibilities of present and foreseeable technology.
I'm not saying its easy. I guess I calibrate my concept of foreseeable technology as sleeker, faster mobile devices being trivially predictable, fusion as possible, and general-purpose nanofactories as speculative.
On that scale, I would place permanent off-world settlements as closer than nanofactories, around the same proximity as fusion. Closer, since no new discoveries are required, only an enormous outpouring of resources into existing technologies.
Well, that for starters.
Then there is the drive to insure the survival and happiness of your children. I have found that this increases with age. If you don't have that drive yet, simply wait. There's a good chance you will be surprised to see that you will develop one, as I have. I imagine this drive undergoes another burst when one's children have children.
Then there is foreclosing on the possibility of the human race reaching the stars. If that doesn't excite you, what does? Sports? Video games? I'm sure those will also spread through the galaxy if we do.
Then there there is the possibility that medical science has a breakthrough or two during your lifetime that makes these more than just theoretical futures.
For the most part, my emphasis is not on limiting population directly. I do believe that charitable efforts have the responsibility to mitigate the risk of a demographic trap in the areas they serve. But I think getting anybody who matters to listen is a lost cause.
My emphasis is on being conscious of the fact that the reason we're still alive and prospering is that we are continuously buying ourselves more time with technology and use this insight to motivate greater investment in research and development. This seems like an easier sell.
Here is a thought experiment that might not be a thought experiment in the foreseeable future:
Grow some neurons in vitro and implant them in a patient. Over time, will that patient's brain recruit those neurons?
If so, the more far-out experiment I earlier proposed becomes a matter of scaling up this experiment. I'd rather be on a more resilient substrate than neurons, but I'll take what I can get.
I'm betting that the answer to this will be "yes", following a similar line of reasoning that Drexler used to defend the plausibility of nanotech: the existence of birds implied the feasibility of aircraft; the existence of ribosomes implies the feasibility of nanotech... neurogenesis occurring during development and over the last few decades found to be possible in adulthood implies the feasibility of replacing damaged brains or augmenting healthy ones.