Forgot your password?

Comment: actually automatic picture caption generator (Score 1) 21

by slew (#48430897) Attached to: Google Announces Image Recognition Advance

Not as "advanced" in image recognition as advertised.

Basically they took the output of a common object classifier and instead of just picking the most likely object (which is what a typical object classifier looks for), it leaves in in a form where multiple objects are detected in various parts of the scene. Then they train a neural network to create captions (by giving it training pictures with associated captions).

According to the paper, it sometimes apparently generates a reasonable description. Other times it reads in picture of a street sign covered with stickers and emits a caption like "refrigerator filled with lots of food and drink".

Actually the most interesting thing about it is the LSTM-based Sentence Generator that is used to generate the caption from the objects. LSTM's are notoriously hard to train and they apparently they borrow some results from language translation techniques to attempt to form intelligible sentences.

This is all very googly-researchy in that they want to see what the limits of pure data driven machine learning are (w/o human tuning). This is not a however much of an advance in image recognition as it is an advance in the language for caption construction.

Comment: Re:Nuclear Test Ban? (Score 1) 449

by slew (#48428915) Attached to: What Would Have Happened If Philae Were Nuclear Powered?

FWIW, NASA has used radio-isotope-thermoelectric generators (aka RTG) in several space missions. RTGs are basically nuclear "batteries". The most high-profile have been Voyager, Cassini, and Curiosity...

The radio-isotopes inside RTGs give off radiation, but it's generally pretty low and all the radio-active parts are sealed inside the "battery" so that even if they blow up, the battery is generally still intact.

The problem with using batteries for propulsion is efficiency. These batteries are heavy vs the amount of power (energy over a short period of time) they can deliver, so you need to use a different way of extracting the energy from the radioactive material which generally is less safe (and more "bomb-like"). That is not allowed.

Thus the RTGs are only used to power the electrical sub-systems and the waste heat is used to keep the electronics at a reasonable temperature in a cold depths of space... kinda like a big battery...

Unfortunately, these RTGs are hard to come by (they are generally powered by a radioactive isotope of plutonium), and the European Space Agency (which launched the probe) doesn't have the ability to make them. Nasa has some, but apparently none were usable for this mission (e.g., if you need AAA and you have only C batteries, you are kinda SOL)

Comment: Re:Mass (Score 1) 449

by slew (#48428723) Attached to: What Would Have Happened If Philae Were Nuclear Powered?

where as the panels afford the craft the possibility of functioning indefinitely...

Except for the small fact that they don't expect the probe to operate as it approaches the sun (it's a comet they are on and they tend to "activate" when they approach the sun).

Had you been talking about a probe set to go well away from the sun then absolutely and pu-238 power plant would be a great idea.

And if they actually did expect the probe to survive the approach to the sun, as it is a comet, it would have definitely gone far-far away from the sun...

AFAIK, one of the main reasons they didn't use and RTG is that no usable ones were available from NASA and the ESA didn't yet have the expertise to make them themselves. Not to say that they still wouldn't have gone solar because of other mission parameters (e.g., mass), but the reasons you give don't really add up.

FWIW, once RTGs are assembled, you don't get to "hibernate" them, they start to decay immediately (kind of like a battery). If you don't have the capability to make them yourself, there might be one on the shelf you can buy from someone else, but it's probably not designed with your power budget in mind and it still has it's original weight, but the amount of power it will put out is reduced by sitting on the shelf. It took 10 years to get to the comet, so that's really puts a lot of constraints on using a pre-existing RTG, perhaps too many for the mission, but if they had the technically ability to make a custom one that matched the mission criteria, I'd bet they would have strongly considered it.

Comment: Re:Ignorant Article (Score 1) 449

by slew (#48428601) Attached to: What Would Have Happened If Philae Were Nuclear Powered?

The fuel is consumed faster than its halflife because of subcritical chainreactions. The amount of chainreactions, and thus the energy output, can be controlled by absorbing neutrons.

Sub-critical chain reactions in RTG are pretty much minimal. The mass and configuration of the radioactive source is selected so that it is essentially a textbook half-life source (from fission cross-section).

The whole point of an RTG is that it doesn't need things like control rods or have failure modes associated with moving parts. Unfortunately, they are generally thermodynamically less efficient generating electricity/energy (relative to waste heat). They also tend to use very heavy encasements for launch safety reasons reducing their weight-to-power efficiency vs solar.

On the more leading edge of RTG research is combining a thermo-electric (e.g., seebeck style) and photo-electric to convert more of the infrared energy that is currently wasted in an RTG to useable electricity. There is also some research into finding so-called high zT materials (ones that have higher conversion efficiency), but results so far have been disappointing (this might be because this research is more niche and thus less lucrative than high temperature superconductors which has occupied researchers for a while). Sadly RTG have very little commercial or military value (Pu is hard to obtain in scale and the it's a niche market to begin with and cannot get bigger) and as a result likely aren't going to get much more attention from researchers.

Comment: Re:and that means it doesn't cost any more? (Score 1) 228

by slew (#48407297) Attached to: The Dutch Village Where Everyone Has Dementia

Easy though it is to harp at the executive pay, it is largely irrelevant to the cost of the final product...

As a first order effect, no, but since executive pay is often tied to the company profitability (either through stock options or bonus plans, or both), a CEO has quite a bit of incentive to massively increase the profitability of the enterprise so he/she (okay, I think they are all he) will reap the percentage rewards (along with all the other stock holders) without regard to the cost of the customers bottom line. This second order effect is simply a natural consequence of how things are set up...

If you think about it, this second order effect would likely a bigger effect than the first order effect... ;^)

Oh yeah, a typical pharma company might actually really need to make insane profit on the home run products in order to cover up for all the crash and burn products they will probably produce, but that doesn't take away from the first argument...

Comment: Re:Anybody familiar with the manufacturing side? (Score 4, Informative) 111

by slew (#48389411) Attached to: An Applied Investigation Into Graphics Card Coil Whine

I understand that high-frequency magnetics are at risk of physical oscillation(the detailed math is right over my head; but all it takes is one part of the part attracting or repelling another part of the part, at least under some input waveforms, and you'll potentially see movement, which easily enough turns to sound); but the seemingly obvious solution is just to pot the magnetics in an adequately thermally conductive epoxy or other encapsulant.

Does anybody know if that just adds too much cost, without performance benefit, and so gets cut during the BOM penny pinching? Do potting compounds have properties that degrade the performance or efficiency of common magnetics? Why is it that, if coil whine is an issue, they aren't just dipping the things in epoxy and calling it a day?

Unfortunately mechanical damping of the inductor vibration isn't as effective as simply reducing the amplitude of driving frequency in the audio bands. Remember this is a sub-harmonic that is being excited by a non-linear coupling to the audio frequency. Basically the energy in a higher frequency is being converted into a lower audible mechanical frequency.

Theoretically, simply changing the mass of the physical oscillation (e.g. cementing it to something heavier) only slightly modifies the frequency of the oscillation (potentially creating more audible noise) and it still doesn't change the energy much. Viscous damping of the mechanical frequency might help a little bit more. Unfortunately, in practice, surrounding things like solder joints in potting compounds is risky as they have a different thermal expansion coefficients and it can cause additional mechanical stress (resulting in reduced mechanical reliability).

In the end, mechanical means are still not going to be as effective as changing the circuit to reduce the amount of switching energy frequencies which are coupled to the audio frequency bands. Probably even from a total system cost point of view...

Comment: Re:If at first you don't succeed... (Score 1) 262

by slew (#48387919) Attached to: Ubisoft Points Finger At AMD For Assassin's Creed Unity Poor Performance

Cheat your customers, cover it up by suppressing reviews, and then lie about whose fault it is.

Now that the video game industry has passed the movie industry, why is it surprising that they use the same entertainment industry tactics?
The more things change... ;^)

Comment: Re:Comcast tried to steal $50 from me (Score 1) 223

by slew (#48383003) Attached to: Overbilled Customer Sues Time Warner Cable For False Advertising

If the people who apply for the rebate get the promised rebate, then how could you possibly claim that anyone is being defrauded?

Fraud is an intentional tort. If they never intend to give the rebate for all eligible people, then it is fraud if they then do not actually do it (even if you don't complain). If not enough money is allocated up front, and if they run out of money to pay all the eligible rebates they receive, then it seems to me to be fraud (although IANAL)...

There are cases that companies when it comes down to the bad publicity and/or the threat of fraud ruling, eventually pony up. But the FTC doesn't really look kindly on that in the US. Even if people eventually get their money, if they don't get it within 30 days, I believe it's considered fraud. Of course you never hear about the cases where there are only a few people defrauded and they don't complain loudly.

As an interesting aside, one famous rebate disaster happened in the UK with Hoover. The cost of complying with the rebate eventually cost the job of the CEO and the company was sold off... Which goes to show, you can defraud a few folks, but when there are too many, you eventually have to pay the piper...

Comment: Re:Comcast tried to steal $50 from me (Score 1) 223

by slew (#48379533) Attached to: Overbilled Customer Sues Time Warner Cable For False Advertising

Rebates are often fraudulent by design. Many companies that offer rebates outsource the rebate to external marketing firms. First, these external marketing firms don't have the same customer privacy code as the original company. Also, marketing companies often buy the rebates on a pooled money basis, they are given a pool of money to distribute to multiple rebate programs and are incentivized to maximize the surplus.

Outsourcing and incentivizing itself isn't fraudulent (just shady), but the reason that it's often fraudulent is that the allocated pool of money to the external marketing firm is never enough to cover the worst case, so they are effectively going into the promotion with the deliberate intent to defraud customers of the rebate and the original company doesn't indemnify the external company for worst-case shortfall (because they don't trust these shady rebate companies enough to think they won't just claim/pocket the money).

Often these small marketing firms often do not have the financial resources to cover the difference (margins are small because they bid down the pool and they don't take out insurance) and often have co-agreements with shady marketing companies that abuse your information. The original companies know this, but generally just look the other way thinking unless it's a total public relations disaster, they will just wipe their hands clean and enlist one of the other available marketing companies in future campaigns.

Comment: Re:We're landing on a comet (Score 1) 74

by slew (#48366589) Attached to: Rosetta's Philae Probe To Land On Comet Tomorrow

I call BS on your BS. Nearly all the stuff that made space flight possible were human achievements on the GROUND.

Most humans in space have been part of nearly ballistic trajectories or computer controlled robots. On many missions, human presence in space was largely for vanity reasons and the missions could have been accomplished with robots. However, there are a few times when humans were key parts of the accomplishment in SPACE which is what I was pointing out. That takes nothing away from other folks fine accomplishments on the GROUND which made it possible.

Comment: Re:We're landing on a comet (Score 1) 74

by slew (#48364347) Attached to: Rosetta's Philae Probe To Land On Comet Tomorrow

I guess there is no real objective measure of what constitutes the peak of human achievement in space. But this has to be up there with the best of them. Go you good thing!

I think one of the candidates for the peak of human achievement in space is the Apollo 11 moon landing done on manual. Or perhaps the first space walk by the USSR...

The comet landing, however, is probably right up there with the other top robotic achievement in space. FWIW, the mars curiosity sky crane one of the other top 10 that comes to mind...

God may be subtle, but he isn't plain mean. -- Albert Einstein