Oleg Zabluda's blog
Thursday, August 09, 2018
 
"""
"""
In one set of experiments, we have determined that the half-life of a lonely neutron is 879.6s. But, in another set of experiments, we’ve found that the neutron has a half-life of 888s (these numbers might be slightly out of date now). The chance of these two being different by accident is now about one in 100,000.

One possible explanation for the difference is that a subset of neutrons decays to a relatively light particle of dark matter. Now, a pair of papers has punctured that proposal.
[...]
The method that finds the shorter half-life counts the number of neutrons in a bottle after an elapsed time. The second experiment counts the number of protons emitted by a beam of neutrons. The beam method would not count neutrons that did not decay to a proton. [...] One possibility is that sometimes neutrons decay into a baryonic dark matter particle. [...] electrically neutral and weakly interacting, we wouldn’t detect it by accident in either experiment.
[...]
To fit this idea into the world around us, we know that the new dark matter particle has to be heavy enough to cope with the observed stability of isotopes [...] and the stability of the proton. [...] Under reasonable circumstances, we might expect neutrons decaying to dark matter to affect the properties of neutron stars. [...] Adding dark matter with mass up to 1.2GeV renders neutron star models incompatible with observations
"""
https://arstechnica.com/science/2018/08/neutron-stars-are-probably-not-hiding-dark-matter-under-their-skirts/
https://arstechnica.com/science/2018/08/neutron-stars-are-probably-not-hiding-dark-matter-under-their-skirts/

Labels:


 
"""
"""
BFR [...] would be the most efficient launch vehicle to LEO ever, besting the latest record holder, the expendable version of the Falcon Heavy launch vehicle (4.49% of launch mass), by over half a percent. [...] thrust to weight ratio of 1.36 to 1, which meant the launch vehicle could potentially lift even more with larger propellant tanks.
[...]
capable of lifting more than 300 tonnes to LEO. [...] due to its sheer size and the high ISP of its engines, barge landing reuse would only cost about 4% of the rocket’s maximum capability.
[...]
SpaceX’s grand ambitions finally became fully apparent when the launch vehicle’s expendable payload capacity of 550 tonnes was announced. The rocket would be 133% more capable than the version SpaceX had planned just the year before.

This incredible performance was enabled by the Raptor engine. Its sea level thrust had increased by 55.5% from 1,961 kN (440,850 lbf) to 3,050 kN (685,667 lbf), its sea level ISP was up 3.9% (334 seconds versus 321.4 seconds prior), it was now capable of throttling 20-100%, and its 300 bar chamber pressure was the highest in rocket engine history.

Its vacuum version would feature an even higher thrust of 3,500 kN (786,831 lbf) and an ISP of 382 seconds, the highest hydrocarbon ISP engine on record, thanks to a very large 200 to one expansion ratio nozzle.
[...]
could deliver up to 450 tonnes of cargo to Mars.
[...]
Each stage would boast a modest delta wing and four landing legs, allowing the stage to de-orbit and land vertically.
"""
https://www.nasaspaceflight.com/2018/08/evolution-big-falcon-rocket
https://www.nasaspaceflight.com/2018/08/evolution-big-falcon-rocket

Labels:


 
"""
"""
On Nov. 8, 2016, Donald Trump was elected as the 45th President of the United States — and Brandon Straka, a gay man and artist living in New York City, posted a video of his reaction to Facebook. “I was devastated. I voted for Hillary, and I was one of those people who was going on social media, crying, making videos,” says Straka.

Almost two years later, Straka posted another video that has since gone viral and spawned a movement. [...] created the #WalkAway campaign in mid-June of 2018, a social media movement that encourages lifelong liberals and Democrats to “walk away” from their party and explore conservative politics with an open mind.

For Straka, the left practices tolerance and diversity in a superficial way, with no regard to individual thought or personal belief: “If you express an opinion that’s outside of what is their ideology, there is no tolerance and there is no diversity.”

“I don’t think that being hostile towards heterosexual people helps gay people,” he says. “I don’t think that being hostile towards men empowers women. I don’t think that being hostile towards white people empowers black people.”

Having grown up in a small town in Nebraska, Straka knew a lot of people who voted for Trump. “I was really on a quest to try and understand why did they vote for this man who was a racist, who was a bigot.” A friend who is a lifelong conservative contacted him, sending a link to a YouTube video titled “Debunking That Trump Mocked the Disabled Reporter.” Straka was skeptical: “I almost still sort of had that liberal rage inside of me, that sort of thought, ‘I can’t wait to watch this and then tell her how stupid she is for being brainwashed by this idiocy.’” The video was a compilation of footage of Trump performing the same flailing hand gestures and rambling voice that he had enacted when imitating a disabled reporter. Brandon was shocked. “It became clear to me that he didn’t mock that man’s disability whatsoever. Yes, the man was disabled, but what he was really doing was making fun of the fact that this person who happened to be disabled was caught in a lie. You know, it blew my mind.”

As Straka dug deeper into the media’s coverage of Trump, he discovered more inconsistencies that disturbed him. “I even found footage where there were groups of black people who went to Trump’s rallies to support him,” he says, “and when [the media] got there, they actually framed up the shot to cut the black people out so that it appeared there were only white people there. … I started to see that I had been incredibly unkind and judgmental towards all of these people in the country who I thought were horrible people, because the media had made me believe that these people were terrible people, when in fact it wasn’t true.”

When he reached out to liberal friends and family, asking if they were aware of how the mainstream media was manipulating the public’s perception of Trump, he found himself shut out. “People were starting to disengage with me, they were starting to cut me off in real life, they were starting to cut me off on social media. There were all these different things that people were saying about me in order to be able to justify that I had sort of walked away from the camp and walked away from the groupthink.”
[...]
In early July, media outlets reported that part if not all of the movement is fictitious, that the #WalkAway campaign created “fake ads” using models to portray supposed members, and that according to the Hamilton 68 site, run by the bipartisan Alliance for Securing Democracy, the hashtag is connected to Russian bots. Straka says it’s not true: “They have always had the option to go to the #WalkAway campaign page on Facebook, look at the real testimonials of real people, but they chose not to do that. No one ever reached out to me for comment, no one ever reached out to me to ask for my perspective as the founder of the campaign.”

Since launching the #WalkAway campaign, Straka has also been accused by people online and in real life of being a member of the “alt-right”: Several weeks ago in a New York City camera store, he was asked by a salesman if he intended to use the items he wanted to purchase for “alt-right purposes.” “I was taken aback,” he recounts, “and I started laughing and I said, ‘Alt-right?’ And he said, ‘Well, aren’t you with the #WalkAway campaign?’ I said, ‘Yeah,’ and he said, ‘Well, I’m sorry, I can’t sell to you.’”
"""
https://www.yahoo.com/news/unfiltered-left-practices-tolerance-superficial-ways-231505016.html
https://www.yahoo.com/news/unfiltered-left-practices-tolerance-superficial-ways-231505016.html

Labels: ,


 
MnasNet: Platform-Aware Neural Architecture Search for Mobile (2018) Mingxing Tan, Bo Chen, Ruoming Pang, Vijay...
MnasNet: Platform-Aware Neural Architecture Search for Mobile (2018) Mingxing Tan, Bo Chen, Ruoming Pang, Vijay Vasudevan, Quoc V. Le
"""
automated neural architecture search approach for designing resource-constrained mobile CNN models. We propose to explicitly incorporate latency information into the main objective so that the search can identify a model that achieves a good trade-off between accuracy and latency. Unlike in previous work, where mobile latency is considered via another, often inaccurate proxy (e.g., FLOPS), in our experiments, we directly measure real-world inference latency by executing the model on a particular platform, e.g., Pixel phones. [...] our approach consistently outperforms state-of-the-art mobile CNN models across multiple vision tasks. On the ImageNet classification task, our model achieves 74.0% top-1 accuracy with 76ms latency on a Pixel phone, which is 1.5x faster than MobileNetV2 (Sandler et al. 2018) and 2.4x faster than NASNet (Zoph et al. 2018) with the same top-1 accuracy. On the COCO object detection task, our model family achieves both higher mAP quality and lower latency than MobileNets.
"""
https://arxiv.org/abs/1807.11626
https://arxiv.org/abs/1807.11626

Labels:


 
Tinder Lies (2018) Irina D. Manta
Tinder Lies (2018) Irina D. Manta
"""
The rise of Internet dating—in recent years especially through the use of mobile-based apps such as Tinder, Bumble, or Hinge—forces us to re-examine an old problem in the law: that of how to handle sexual fraud. Many people with romantic aspirations today meet individuals with whom they do not share friends or acquaintances, which allows predators to spin tales as to their true identities and engage in sexual relations through the use of deceit. Indeed, according to some studies, about 80% of individuals lie on at least some part of their online dating profiles, and a subset of those individuals tell lies that undermine the foundation of their sexual mates’ subsequent ability to give consent. Whether or how to criminalize this type of fraudulent behavior has been debated for some time, and the difficulties involved in prosecutions in this context have made the criminal law a fairly ineffective tool. Previous proposals for tort recovery have failed to gain many adherents for similar reasons. This Article seeks to strike a new path by proposing, first, that we harness the tools of trademark law to reduce search costs and deception in the dating marketplace just like we do in the economic marketplace. Second, it argues that we should use a streamlined process through small claims courts to discourage behaviors that may bring significant dignitary, emotional, and other harms to people’s lives. Last, it proposes the use of statutory damages to alleviate the difficulties in accurately gauging the remedy level for the harm from a given instance of sexual fraud. By providing recovery in cases of material lies like trademark law does in cases involving deceptive marks, this Article takes an important step toward aligning the legal framework of sexual fraud with those of other types of misrepresentation, incentivizing transparency in the increasingly murky dating world, and protecting individuals’ ability meaningfully to consent to sexual relations.
"""
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3229223
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3229223

Labels:



Powered by Blogger