cultus

42 ms·

421.
by cultus 8y ago
I notice this doesn't mention hinge loss, which is by far the simpler way of arriving at the SVM. Hinge loss is just max(0, 1- t*y), where y is the output of the linear model and t = +-1 is the label. Thus, it takes the common-sense ap
422.
by cultus 8y ago
There's been some work on variational Bayesian formulations of SVMs in the last few years. These can give actual uncertainty estimates and do automatic hyperparameter tuning. This one in particular is very cool: https://arxi
423.
by cultus 8y ago
Yeah, that's probably more important than the weather. Their cities are incredibly walkable. I recall reading that in the US, folks living in suburban areas average something like 3 or 4 kilos more than those in urban areas.
424.
by cultus 8y ago
It's probably moreso the Mediterranean diet and the Mediterranean climate (encouraging more walking). The evidence that wine or any alcohol consumption positively affects health is equivocal at best (I say as I drink an IPA).
425.
by cultus 8y ago
No, it's not. The effect size that purportedly occurs according to this research (double-digit percentages of cases caused by toxo) would have been detectable on a population level. There would be marked differences in schizophrenia ra
426.
by cultus 8y ago
These studies have usually showed a large effect size and a small sample. That doesn't explain it.
427.
by cultus 8y ago
It's no surpise that it should be herdlike, since the algorithms are operating with much of the same information.
428.
by cultus 8y ago
This is just p-hacking. It's really sad that in 2018 researchers are still pumping out garbage like this. All of these toxo behavior studies have been very poor quality. For example, there's been a few studies purporting that folk
429.
by cultus 8y ago
No, they did not get answered by Slat in the comments. Are we reading the same article? Dr. Kim Martini is a very respected oceanographer. While you may not like the tone of the article, she knows what she is talking about. Her concerns wer
430.
by cultus 8y ago
This project cannot work. Here's some information written by an actual oceanographer: http://www.deepseanews.com/2014/07/the-ocean-cleanup-part-2-... http://www.deepseanews.com/2016/06&#x
431.
by cultus 8y ago
I'd rather not go into details on the sexist attitudes, but let's just say there is a pattern of talking down or ignoring the credentials of female scientists. The feasibility study is BS. It only convinced the media, no one in th
432.
by cultus 8y ago
The feasibility study was, to be blunt, horseshit. As I stated at the top, they are very good at getting the media to buy their bogus claims. Here are some articles written by actual marine scientists and engineers regarding this project:
433.
by cultus 8y ago
Well, it's not a significant amount of new plastic on a global scale. However, why does it make sense to spend gobs of money on something that at best won't help and at worst will make things slightly worse? Why not invest in prov
434.
by cultus 8y ago
It's not a significant amount. It's important to grasp just how huge the ocean is. This would clean up a pointlessly small fraction of the garbage patch. Not only that, but it would be destroyed in a storm quite quickly, generatin
435.
by cultus 8y ago
I would agree that it is unfair, except as I point out they deliberately ignore or attack critics, and do not offer any actual valid scientific arugments in return. This pattern includes ignoring or even demeaning scientists and engineers
436.
by cultus 8y ago
Generally, poorer countries tend to be bigger sources due to worse sanitation. A huge proportion is from stryofoam or other plastics on fishing nets or buoys. Plastic bags are the single biggest source on land. Plastic bag bans are really i
437.
by cultus 8y ago
The Ocean Cleanup Project is essentially a scam at this point. It can't work for numerous ecological (scoops up ocean life) and engineering (the ocean is HUGE, corrosive, and violent) reasons. Not to mention that ocean microplastics ar
438.
by cultus 8y ago
White tea has such little caffeine because it is steeped for only a few seconds at a low temperature. Again, it's the difference in preparation methods rather than the leaf.
439.
by cultus 8y ago
Typical LA courses in math departments have a bizarre focus on being able to do Gaussian elimination by hand and stuff like that. It's not particularly useful or even mathematically interesting. LA courses would be so much more useful
440.
by cultus 8y ago
Green teas have about the same amount of caffeine. If anything, they'd have more since they are less processed. But as prepared, they have a lot less because they are steeped for shorter periods of time and with cooler water.
441.
by cultus 8y ago
Getting rid of CFCs is orders of magnitude less painful than fossil fuels, concrete, and habitat destruction.
442.
by cultus 8y ago
For this reason, a long term goal of mine is to move to New Zealand. In retrospect, we probably had a more likely chance than not of the Cold War turning nuclear. There were several extremely close calls averted by dumb luck. In the next Co
443.
by cultus 8y ago
>It's striking how many academics admit (usually anonymously, online) to using Sci-Hub for content they do have legitimate access to. Sci-Hub appears to be pretty much flawless at looking up articles across all publishers by ID or e
444.
by cultus 8y ago
The fact that it is illegal makes it a problem to use for many people I know, especially in government. That in the TLD changes like every week.
445.
by cultus 8y ago
Here's a fantastic set of cartoons explaining concurrency and parallelism: https://typelevel.org/cats-effect/concurrency/basics.html
446.
by cultus 8y ago
But yet it is the most common type of bug, even in production. Clearly, it's not that easy to deal with for human programmers. What's the problem with letting the compiler help you?
447.
by cultus 8y ago
If it is a requirement for something not to be null, unwrap the option before you send to it to the part of the program that can't accept nulls, and deal with the case of None in a sane way and in a predetermined place. Then you don&#x
448.
by cultus 8y ago
>The most trivial pointer issue is a NULL pointer. This is such a trivial issue to catch its hardly even an error, yet people use that case as the exemplar for NULL issues. How can you claim that NPEs are "hardly ever an error.&quot
449.
by cultus 8y ago
Replying to your lower comment (the coffee has kicked in): The situation you describe is one where a null really is an unrecoverable error, and the program should terminate. That is the one case where it makes sense to just let a NPE happen
450.
by cultus 8y ago
>This is not the source of the vast majority of pointer errors. >Checking for (and trapping) NULL pointer dereferences is trivial, what is more difficult is the rest of the pointer range that doesn't get checked but is equally in

More ›