Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror

Comment: Re:Established science CANNOT BE QUESTIONED! (Score 2) 590

by Entrope (#48634423) Attached to: Skeptics Would Like Media To Stop Calling Science Deniers 'Skeptics'

Which people do you think I am describing? There certainly are a lot of weirdo extremists in the environmental-activist camp, but I wasn't really thinking about them. If you want me to ignore the weirdo extremists on that side, will you ignore the weirdo extremists on the other side? More significantly, will media and activists stop focusing on the (conveniently distracting) anti-AGW weirdo extremists so that we can pay more attention to what actually can and should be done?

What specific steps do the reasoned thinkers recommend as "what actually needs to be done"? Last I heard, European countries were revising or just rolling back climate agreements because (a) they realized they couldn't achieve their goals without reducing their quality of life, (b) they realized the system was being gamed, and/or (c) they wanted to keep up with the countries who didn't sign up to those agreements.

Comment: Re:Established science CANNOT BE QUESTIONED! (Score 4, Insightful) 590

by Entrope (#48633595) Attached to: Skeptics Would Like Media To Stop Calling Science Deniers 'Skeptics'

Lots of people believe in ghosts. Lots of people also believe in people who "think[] that human activities have no impact on climate change". There's about as much hard evidence in one of these beliefs as in the other.

When climate alarmists stop pretending that the dispute is over the degree of human influence on climate, and how much different countries should spend to mitigate anthropogenic climate change (or other kinds!), they might start to get traction with skeptics. Also when they start acting like the situation is as bad as they claim it is.

I know that when I used an electric sous vide cooker to make pork chops for dinner last night, it was worse for the climate than if I ate raw vegetables, and better than if I grilled a slab of steak over a bonfire. I know that living in the suburbs emits more greenhouse gases than living in a tiny apartment in a big city. I am thoroughly unconvinced that forcing most people to live like the alarmists claim we should (but usually don't live themselves) will yield the claimed benefits, or be worth the costs even if the benefits would be as claimed.

Comment: Re:Clickbait (Score 1) 125

by Entrope (#48628355) Attached to: Research Highlights How AI Sees and How It Knows What It's Looking At

I called it cheating because they violated both one of the prime rules of AI: train on a data set that is more or less representative of the data set you will test with, and one of the prime rules of statistics: do not apply a priori statistical analysis when you iterate with feedback based on the thing you estimated. Their test images are intentionally much different from the training images, which is one of the first things an undergraduate course on AI will talk about. They also use what are essentially a priori estimates after they repeatedly tweak the inputs to push those estimates to extremes, which is identified as taboo in decent undergraduate courses on statistics. Both of those are intentional violations of good practices that make the results look worse for the neural networks.

I can't tell from their paper what they mean by "99% confidence". Unless the DNN has max-pooling layers very near the output, none or many of the output units might have high activation levels for a given input. (It sounds like they had classes with low typical activation levels, and did not try to evolve fooling images for those classes.) If that happens -- say, "wheel" gets a score of 0.99, "lizard" gets 0.90, "dog" gets 0.80, and everything else is near zero -- then it is inappropriate to say that the network decided it was a wheel with 99% certainty. You would usually say that the network recognized the image as a wheel, but note it as an ambiguous result.

Comment: Re:Clickbait (Score 1) 125

by Entrope (#48625157) Attached to: Research Highlights How AI Sees and How It Knows What It's Looking At

Why was my characterization of their approach "hardly fair"? Someone -- either the researchers or their press people -- decided to hype it as finding a general failing in DNNs (or "AI" as a whole). The failure mode is known, and their particular failure modes are tailored to one particular network (rather than even just one training set). I think the "hardly fair" part is the original hyperbole, and my response is perfectly appropriate to that. The research is not at all what it is sold as.

Don't multi-class identification networks typically have independent output ANNs, so that several can have high scores? I assumed, perhaps incorrectly, that the 99+% measures they cited were cases where only one output class had a high score, and the rest were low. If they were effectively using single-class identifiers, either in fact or by considering only the maximum score in a multi-class identifier, that makes their findings even less notable.

Comment: Re:Clickbait (Score 1) 125

by Entrope (#48624757) Attached to: Research Highlights How AI Sees and How It Knows What It's Looking At

The researchers also basically cheated by "training" their distractor images on a fixed neural network. People have known for decades that a fixed/known neural network is easy to fool; what varies is exactly how you can fool it. The only novel finding here is their method for finding images that fool DNNs in practice -- but the chances are overwhelmingly high that a different DNN, trained on the same training set, would not make the same mistake (and perhaps not make any mistake, by assigning a low probability for all classes). It is a useful reminder for some security analyses, but not a useful indictment of AI or DNNs as a whole.

Comment: Re:This synopsis (Score 1) 125

by Entrope (#48624703) Attached to: Research Highlights How AI Sees and How It Knows What It's Looking At

The people choosing the training sets are not morons at all. This "research" is almost exactly analogous to finding that this year's SAT can be passed by feeding it a fixed pattern of A, C, D, A, B, and so forth -- and then declaring that this means standardized testing is easy to fake out. They are exploiting the particular structure of a particular instance of a DNN. It is not surprising that they can find odd images that make a DNN answer "yes" when the only question it knows how to answer is "is this a rotary phone dial?"

Comment: Re:Displacing five times as much water... (Score 1) 113

by Martin Blank (#48623697) Attached to: New Cargo Ship Is 488 Meters Long

You would think with that volume of gas you would be up there with a nuclear sized detonation.

It has a capacity of some 430 million liters of LNG. At an average density of 0.463 kg/L at -160C, that's 199 million kilos of liquefied methane. At 22.2 MJ/L, that's 4.42 billion MJ, or a shade over a megaton of TNT if it were to all go off at once.

Though I doubt that's possible. The storage facilities will have separation, so at best there would be a chain that would dampen the impact somewhat.

Comment: Re:class act (Score 1) 171

by Martin Blank (#48579805) Attached to: Julian Assange Trying To Raise Nearly $200k For a Statue of Himself

It's not quite nothing--he did retweet it to give it some attention--but I thought it was iffy myself, and I am certainly no fan of Assange. I keep him on one of my Twitter lists just because his delusions amuse me (and because he sometimes posts something interesting). When something this unusual pops up, it's best to look into it a bit further.

Comment: Re:Every 30 days. (Score 1) 247

by Martin Blank (#48540857) Attached to: Ask Slashdot: Convincing My Company To Stop Using Passwords?

I use sentences of my own creation. In the case of mandatory password changes, I will sometimes use some piece of trivia. For example, I might use the counties of a state. It reduces the entropy somewhat, especially if someone finds out what the reference is, but it allows me some room to work and embeds a new bit of trivia into my head.

I do use password managers (a couple of them, actually), and I know there are some enterprise password managers out there. There's a danger to stand-alone managers, but a well-managed enterprise should have all of the core passwords securely stored somewhere.

Comment: Re: rounding error (Score 1) 71

by Martin Blank (#48529157) Attached to: Technical Hitches Delay Orion Capsule's First Launch

More launches mean more cost, especially if you're scattering it across launch pads located around the world. There aren't many sites that can handle significant launch masses: Cape Canaveral, Baikonur, Plesetsk, French Guiana, Jiuquan (China), Satish Dhawan (India), and Tanegashima (Japan). So you have enormous coordination between nations that have widely varying launch experience for their heavy lifters, that use different technologies and procedures, and have different goals for their space programs. This doesn't even get into the politics of "What do you do for me if I agree to lift this 15T payload into orbit?"

It also would cost more fuel, since launching from different locations means having to match inclinations. This has already led to one major limitation with the ISS, since its inclination is a compromise between the ideal inclinations for Cape Canaveral and Baikonur.

On top of that, you add complexity in having to dock so many more times, increasing the risk of an incident. While the potential loss from a single large launch is significantly more than that of a single small launch, the cumulative risk of any loss is greater with multiple launches. Putting a thousand tons into orbit would take eight SLS launches, but a minimum of 44 launches of the Delta IV Heavy or Proton, currently the heaviest launchers available.

I would rather see projects like the Falcon XX or MCT encouraged, and I expect they'll be showing up on the test schedule around the same time as the SLS. But NASA is going to have their own path despite the costs, and so they may as well work on an SLS-class launcher. If nothing else, it will give SpaceX (and maybe others) something to aim for and probably provide some valuable lessons along the way.

Comment: Re:Over what time interval? (Score 1) 528

by Martin Blank (#48528237) Attached to: The Sony Pictures Hack Was Even Worse Than Everyone Thought

For one thing, that is likely the storage size, not the transfer size which is likely going to be way less due to compression.

The transfer size probably is smaller to some degree. But to hit that uncompressed volume of storage size, there is going to be a lot of data with poor compression rates. I expect that a lot of pristine, high-resolution digital video is in that, and that certainly won't compress all that well.

But as you point out, those can be terabytes in size. Even with the potential value of that, most people aren't going to download the raw files, and fewer still will go through the work of converting them to lower-res files more amenable to download. I'm not saying it won't happen, just that I think it's unlikely. Sony has more to worry about from the financial and personal information that was obtained than the revenue loss from any movies that were downloaded.

Our policy is, when in doubt, do the right thing. -- Roy L. Ash, ex-president, Litton Industries

Working...