argonaut

66 ms·

721.
by argonaut 10y ago
To reiterate, people already widely knew multiple layers solved XOR at the time.
722.
by argonaut 10y ago
This is not true. We can look at networks and manually look at their weights to determine what features it learned, even high level features. You can do it right now (by examining a pretrained network). People don't bother to because
723.
by argonaut 10y ago
Sure. My other point was that these are not high school graduates going to bootcamps. These are people with jobs and savings who are transitioning careers.
724.
by argonaut 10y ago
To be clear, those solutions were well known at the time. It was known at the time that multiple layers could compute any boolean fucntion ( https://en.wikipedia.org/wiki/Perceptrons_(book) ). The role of Minsky in killi
725.
by argonaut 10y ago
You were implicitly asserting that Ross Ulbricht is popular with the public, as well. I was pointing out how ridiculous that is.
726.
by argonaut 10y ago
FWIW, it appears many of these concerns were known and brought up during the original scandal 30 years ago: http://www.nytimes.com/1989/05/21/world/trial-in-spain-on-to... http://www.nytimes.c
727.
by argonaut 10y ago
Many people also find ways to pay for college without debt. Furthermore, most people going through these bootcamps are doing it for a career transition. It's nontrivial but totally doable to save $20k for someone working.
728.
by argonaut 10y ago
I was just pulling a number out of thin air. We're in scifi speculation territory so the numbers are all made up.
729.
by argonaut 10y ago
News flash: if you polled the average American about Silk Road and the illicit things sold there, they'd be in favor of shutting it down.
730.
by argonaut 10y ago
Most colleges cost far, far more, and yet many many people find a way to pay for them.
731.
by argonaut 10y ago
Being incredibly popular with consumers and the public will do that for you.
732.
by argonaut 10y ago
There is no reason to believe it is even possible to even create such a God-like intelligence. Even harnessing the energy of the sun would not give you enough energy to compute anything about the universe . We also have every reason to bel
733.
by argonaut 10y ago
Fresh grads. Entry-level SWE.
734.
by argonaut 10y ago
This is generally overblown. What has mainly happened so far is that already-existing products have been incrementally improved by deep learning - speech recognition, spam detection, search ranking, etc. all worked reasonably well before de
735.
by argonaut 10y ago
The brain does not really have difficulty with camera angles. Almost every single optical illusion / visual trick / visual quirk is actually an example of how incredibly clever and well-optimized the brain is. The fact is the brai
736.
by argonaut 10y ago
You need to adjust your numbers. Fresh grads at Google/Facebook make >160k per year, total compensation.
737.
by argonaut 10y ago
Are you kidding? Advertisers trip over themselves for the chance to advertise to consumerism-drenched, trend-driven teenagers and soon-to-be twenty-somethings.
738.
by argonaut 10y ago
Your numbers are way off. Fresh grads (Bachelors degree) working at Google/Facebook are now making > $160k total compensation.
739.
by argonaut 10y ago
This seems like a very big, pretty unreasonable assumption: "Assuming that a superintelligence will contain a program that includes all the programs that can be executed by a universal Turing machine on input potentially as complex as
740.
by argonaut 10y ago
Seed is roughly the same as it's always been. Even if a lot of VC firms are retrenching, there are still tons of angel investors and seed funds out there casting wide nets. Series A, definitely.
741.
by argonaut 10y ago
It seems everyone is a forensic medical professional now. Did anyone actually read the report? From the medical examiner: "I received and reviewed the Toxicology Report. The findings do not change my opinion regarding the cause and man
742.
by argonaut 10y ago
No, it's not the same. Not sure why you think that analogy even makes sense.
743.
by argonaut 10y ago
Yes, but that's the problem. Just "letting it do its thing" and relaxing/trusting/ expecting it to "do the right thing" is what leads to accidents like this. And if you trust Autopilot to do the right thi
744.
by argonaut 10y ago
Part of my criticism of Tesla, though, comes from the fact that a large truck crossing in front of you is actually not a particularly complex situation for humans (just brake), even for a sleepy driver.
745.
by argonaut 10y ago
Not trolling. Academics routinely solve huge challenges in research fields. No reason a retired engineer can't solve problems in a niche industry field.
746.
by argonaut 10y ago
Are you an academic? Because peer review is almost always (at least in CS) blind, with authors anonymous to the reviewers.
747.
by argonaut 10y ago
Yes, but this doesn't add much because any human situation involving more than 2 people involves interpersonal dynamics, AKA politics. It's a question of how much .
748.
by argonaut 10y ago
I'm not an academic, but I think you'll find many academics would be against this. Things would get more complicated because there would obviously need to be a vetting process on who could comment. Things shouldn't be a popul
749.
by argonaut 10y ago
I rarely communicate with other drivers (other than with signal lights and brake lights), so I don't actually think it's necessary for self-driving cars. The only use of communication is right of way at stop signs, but this can
750.
by argonaut 10y ago
Yet a one-person professor research team routinely solves big challenges in certain research fields?

More ›