If brute force is more effective... then it's better. There's probably a post in the sequences about this. Do you care about machine vision, or do you care about fancy algorithms? If you care about machine vision, then you should want the most effective approach (the one that is best at "winning"), whatever its nature. There is no such thing as "cheating" in pursuit of such a goal. On the other hand, if you care about fancy algorithms (which is a legitimate thing to care about!), then brute force is by definition uninteresting.

I don't think that it's quite this easy to depict the problem. In applications where I think the brute force SVM approach is fundamentally a more correct way to model and attack the problem, I'm all for its use. At the same time though, I don't care at all about "fancy algorithms." What I think is that the landscape of modern research is far too risk averse (there is a book coming out soon by Peter Thiel and Garry Kasparov of all people on this very topic -- that human ingenuity and insight has actually slowed in the last few decades despite advances in computing technology).

I think that to solve hard machine vision problems, like perception and cognition, you have to depart from the machine learning paradigm. Machine learning might be a low-level tool for dealing with feature extraction and selection, which plays a role, but is far from effective for higher end problems. For example, I work on activity and behavior understanding. On any timescale above about 1-2 seconds, humans consistently recognize activities by projecting protagonist style goal structures onto their observations (i.e. the cue ball is trying to hit the black ball into that pocket... the cue ball becomes an agent with volition instead of merely a pawn in some sort of physics experiment). Currently, machine vision researchers approach the task of activity understanding as they approach everything else, just train a giant SVM (or some other kernel classifier, sparse coding, dictionary methods, etc. etc. etc.) The performance is state of the art for specific applications, but we don't actually know anything about how to perceive activities. It offers no insight whatsoever into the real cognitive structure that's underlying the task of perception. There are obviously many controversial opinions about a question like this, but my perspective is that probabilistic reasoning and graphical models are a much better way to work on this sort of problem, but even those methods need to be researched and extended to a much more theoretically mature level than where they currently are.

But no one (and I mean no one) will pay you money to do that style of research. Unless you produce a widget that performs task X for company Y at state of the art levels and can deliver it very soon, you get no funding and you get shunned by the major computer vision journals (ECCV, ICCV, CVPR, NIPS, SIGGRAPH). To get your work published and to win grants, you have to market it based primarily on the way you can promise to deliver pretty pictures in the short term. Machine learning is much better at this than advanced mathematical abstractions, and so advanced mathematical abstraction approaches to vision (which will benefit us all very much further into the future) are not being funded by risk averse funding sources.

Nobody is demanding that machine vision must succeed at preposterously difficult tasks anymore. Most consumers of this sort of research are saying that computing power is sufficient now that, as far as commercial applications go, we just need to hammer away on efficiency and performance in very controlled, specific settings. Being a "general vision theorist" has utterly no place anymore, in academia or in corporate research.

This is the root of my specific issue. My goal is to study difficult problem in computer vision that cannot be well-solved with the machine learning paradigm. I believe these need advanced theory to be well-solved, theory which doesn't exist yet. Just like the A.I. winter though, someone would have to fund me without knowing that the end result will benefit them more than their competitors, or if there will be any commercial benefit at all. My goal is to understand perception better and to only start to worry about what that better understanding will do for me later on.

Also, can you be more specific about how this works at Wolfram Research. I frequently attend their career presentations here at my university and then try to talk to the technical representatives afterward and learn about specific opportunities. It doesn't seem to be related to the kind of tool-creating research I'm talking about. In fact, I think that proprietary programming platforms like Matlab, Mathematica, and Maple, are a large part of the problem in some respects. These tools teach engineers how to be bad programmers and how to care more about hacks that produce pretty pictures than about really insightful methods that solve a problem in a way that provides more knowledge and more insight than merely pretty pictures.

When I mentioned "creating tools that go in the toolbox of an engineering" what I mean was inventing new techniques in say, calculus of variations, or new strategies for simulated annealing and Markov random field analysis. I mean theoretical tools that offer fundamentally more insightful ways to do modeling for engineering problems. I did not mean that I want to create software tools that make easier interfaces for doing machine learning, etc.

The default Singularity scenario is probably a bad one, and most scientific/technological progress just brings the Singularity closer without making a positive scenario more likely.

How much of modern science brings one closer to a potential intelligence explosion type Singularity event? If such an event is something that is likely to occur it would need to not be dependent on a lot of different technologies.

So what technologies could actively be a problem?

Well one obvious one is faster computers. The nightmare scenario is that we find some little clever trick we're missing to run smart AI and the first one we turn on thinks hundreds of times faster than us at the start.

The next possible really bad set of technologies are nanotech stuff. If the AI finds an easy way to get access to highly flexible nanotech based on methods we have then we're sort of screwed. (This one seems extremely unlikely to me. The vast majority of people in nanotech keep emphasizing how difficult any sort of constructor bot would be.) The next issue is possible advanced mathematical algorithms. The really bad case here is that an AI gets to look at the arXiv and quickly sees a set of papers which when put together give something like a general SAT solver that solves 3-SAT with n conditionals in Kn^2 steps for some really small constant K. This is bad.

Similar remarks apply to an AI that finds a really fast quantum algorithm to effectively solve some NP hard problem. Seriously, one of the worst possible ideas you can have in the world is to run an AI on a functioning quantum computer or give it access to one. Please don't do this. I'm someone who considers fooming-AI to be unlikely, and believe that BQP is a proper subset of NP and this possibility makes me want to run out and scream at people like Roger Penrose who specifically want to see if we need a quantum computer for intelligence to work. Let's not test this.

But outside these four possibilities the remaining issues are all more exotic and less likely. For example, I'm not worried that an AI will right out of the box figure out a way to make small wormholes and take advantage of closed-time like curves simply because if it has that sort of tech level then it has already won.

So the vast majority of scientific research seems to do very little for helping an AI go foom.

But it does seem that continued scientific research does help us understand which sort of AI threats are more likely. For example if we end up proving some very strong version of P!=NP, then this will make the clever algorithms attack much less likely. If BQP is strictly less than NP in a strong sense, and room temperature strong nanotech turns out to be not doable then most of the nasty foom scenarios go away. Similarly improving computer security directly reduces the chance that an AI will manage to get access to internet things it shouldn't (although again, basic sanity says anything like a strong AI should not be turned on with internet access, so if it gets internet access it has possibly already won. This is reduces the chance of a problem in one specific scenario that isn't terribly likely but it does reduce it.)

Furthermore, scientific progress helps us deal with other existential risks as well as get more of a handle on which existential risks are a problem. Astronomy, astrophysics and astrobiology all help us get a better handle on whether the great filter lies behind us or in front of us and what the main causes are. It wouldn't for example surprise me if in 30 or 40 years we will have good enough telescopes that we can not only see Earth like planets we can see if they had massive nuclear wars (indicating that that might be a possible major filtration event) or that the planet's surface is somehow covered with something like diamond (indicating some point possibly in the very far past, a serious nanotech disaster occurred). A better space program also helps deal with astronomical existential risks like asteroids.

So, overall it seems that most science is neutral to a Singularity situation. Of the remainder some might increase the chance of a near term Singularity and some might decrease it. A lot of science though helps deal with other existential risks and associated problems.

I can address the other questions later on, but I am actually interested in looking to complexity limits for FAI problems. My initial reaction to Yudkowsky's post about cohesive extrapolated volition was that such a thing is probably not efficiently computable, and even if it is, it is probably not stable (in the control theory sense; i.e. a tiny error in CEV yields a disastrously large error in terms of the eventual outcome). It isn't like there is just one single time that we have to have a mathematically comprehensible description of volition. As computational resources grow, I imagine the problem of CEV will be faced many times in a row on rapidly larger scales, and I'm interested in knowing how a reasonable CEV computation scales asymptotically in the size of the projected future generation's computing capabilities. Very very naively, for example, let's say that the number of processors N of some future AI system plays a major role in the mathematical structure of my description of my volition that I need to be prepared to hand to it to convince it to help me along (I know this is a shortsighted way of looking at it, but it illustrates the point). How does the calculation of CEV grow with N. If computing the CEV in a mathematically comprehensible way grows faster than my compute power, then even if I can create the initial CEV, somewhere down the chain I won't be able to. Similarly, if CEV is viewed as a set of control instructions, then above all it has to be stable. If mis-specifying CEV by a tiny percentage yields a dramatically bad outcome, then the whole problem of friendliness may itself be moot. It may be intrinsically unstable.

As far as "math teaching at a respected research university" goes, there are a few reasons. I have a high aesthetic preference for both mathematics and the human light-bulb-going-off effect when students overcome mathematical difficulties, so the job feels very rewarding to me without needing to offer me much in the way of money. I enjoy creating tools that can be used constructively to accomplish things, but I don't enjoy being confined to a desk and needing to focus on a computer screen. The most rewarding experience I have found along these lines is developing novel applied mathematical tools that can then be leveraged by engineers and scientists who have less aversion to code writing. Moreover, I have found that I function much better in environments where there is a vigorous pace to publishing work. At slower places, I tend to chameleonize and become slower myself, but at vibrant, fast-paced places, I seem to function on all cylinders, so to speak. This is why a "respected research university" is much more appealing than a community college or smaller state level college.

I'm very disillusioned with the incentive scheme for academia as a whole. Applied mathematics with an emphasis on theoretical tools is one domain where a lot of the negative aspects have been kept at bay. Unfortunately, it's also a field where statistically it is very hard to get a reasonably stable job. As far as areas of math go, I greatly enjoy theoretical computer science, probability theory, and continuous math that's useful for signal processing (complex analysis, Fourier series, functional analysis, machine learning, etc.)

I had not seen the previous post on career choice and will look into it. But the main reason for this thread was that I think that as far as getting a job and sustaining myself goes, I'm better off trying to hack my preferences and causing myself to actually enjoy computer programming, instead of finding it loathsome as I do now. This is based on a non-trivial amount of interaction with people in the start-up community, in academia, and at government research labs.

there is a trend for some Silicon Valley types to invest in long-term basic research that universities don't support.

Can you point me to any specific examples of this? I have a grad student colleague here who is very involved with face detection and tracking and his work has essentially blown the state-of-the-art performance out of the water. Because of this, he's heavily involved with various startups and web businesses looking to use his better face detection methods. When I queried him for advice, he basically said that not only is long-term, basic research very risky (especially if the researcher has a tendency to look for elegant mathematical solutions), but literally no one will pay you for it. He insisted that you won't find any companies doing long term basic research because it won't benefit them more than competitors in the long run.

One counter example to this might be Willow Garage. However, I think they still are not doing the very basic theoretical math research that they will actually wish they had once a personal robotics industry does start booming. I've really racked my brain trying to come up with places that actually pursue the theory because of the long term practical benefits.

Moreover, I am very discouraged about the state of academic publishing right now. The main reason I want to hack myself is to change my preferences about being a university researcher. Currently, I see only two alternatives: university researcher or corporate/industrial/government researcher. I had always thought that in the former, people paid you grant money because of your ingenuity and foresight and the whole point was to use tax money to allow researchers to conduct high-risk research that commercial entities could not afford to risk their money on. As it turns out though, both of these options require you to pander to whatever the popular commercial interests of your day happen to be, even if you think that the popular commercial interests have got it all wrong and are going down the wrong track.

It makes me feel that I need to hack myself to want to want to just be a programmer for some company somewhere. Make enough money to do cryonics and have an internet connection and just float along. I feel very discouraged to really try anything else. And since my current preferences hold that computer programming for the sake of widget building is soul-crushingly terrible, I feel like it's a big Catch-22 putting me in an ambivalent stalemate.

Lastly, the altruism argument you mention doesn't appeal to me. I think society should have a tenure class of professors able (and required) to do riskier / theoretical research. The amount of work it takes to get to that position in life ought to outweigh whatever solely-free-market altruism a corporate scientist might be prideful of. But the reality is that by the time I am in a position to seriously apply for a tenure track job (2 more years of school, 2 years of post-doc, 5 years as assistant professors, so roughly 9 years from now), tenured positions just simply will not exist. It's already a dying business model, which makes me feel like all of the time I've already spent training non-practical skills into myself was a massive unforeseen waste.