Thanks for sharing.
853 karma · joined January 20, 2020
https://www.ams.org/journals/bull/2022-59-01/S0273-0979-2021...
> Deutsch, too, seemed impatient with my dissatisfaction over our understanding of the nature of reality. “Someone might equally well say: We may know what dogs look like and how they behave, but we don’t know what a dog ‘actually is’,”
You sit further away from your screen (relative to your screen size), than the average user. Let’s work backwards:
Assume you use 150% zoom on average. And you sit 100cm from computer screen. Then I predict an average user would sit 100cm/1.5 =about 66cm from same screen.
edit: or a screen with the same PPI or pixel density
So The Cloud, arguably the peak of human technology, uses contributes a tiny fraction of our total carbon emissions. Is curbing this technology in in any major way a productive way to reduce our emissions?
These emissions are all electricity usage converted to carbon emissions. Would it not be more productive to focus on transitioning all power generation to solar, wind, hydro, nuclear, etc?
Since the calculation involves taking log_2, even if the estimate for the "used power" is off by a factor of 4 the result will only change by 50 years.
Can you make your qualitative comment quantitative and update the numbers in my doubling thought experiment?
I would be curious about what you think is more realistic.
The experiment produced
> 11 megawatts of power
and at Jet
> two 500 megawatt flywheels are used to run the experiments
log_2(2*500/11)*(2022-1997) + 2022 = 2185
At this rate we will have fusion by 2185 I guess?
“Yandex, Mail.ru Group, and Rambler will deindex domains that have 100 links to pirated content starting September 2022”
Can anyone find the preprint? This blog post was from April 2021 so it might be available somewhere, but I cannot find it.
I use linux and system link the home folder to a Dropbox folder. There is an extended history option [1], so that you can go back a year for any file. I think it’s a reasonably close solution if you do not mind the cost and syncing to cloud.
[1]https://help.dropbox.com/files-folders/restore-delete/extend...
I think the answer must be related to an effort to reduce the amount of unauthorized copyrighted material or fraud on the platform. Currently the majority of such material/fraud on youtube is uploaded by small channels who can create another channel as necessary. Throttling small accounts is one possible way to combat this.
Best for whom? [1]
[1] https://www.nytimes.com/2022/01/28/dining/restaurant-phone-l...
You could self host, but setting it up is a time investment and probably not worth it unless you have fun doing it
> Terence Tao on 27 January, 2022 at 10:31 am The production values of the lecture series given by Masterclass would not be possible for me to replicate by myself. (For instance, the film shoot alone involved well over a dozen people, including makeup, prop design, camera and lighting crew, director, editing, etc., in a custom studio with professional-quality recording equipment. The professional scriptwriting services were also very valuable, improving the selection and arrangement of material in ways I would not have thought of.)
> you can Download the JSON Proof file, and share that publicly. It does not reveal any information of the coordinates you discovered.
But it is not at all clear that the proof does not reveal any information of the coordinates you discovered.
Similarly, it is also unclear that the JSON actually proves that you found the treasure. For example, there could be a specific string in the JSON that the program is checking
It is a cool example of coding, but not of zero knowledge proof in my opinion
So a student makes an arithmetic error and they author concludes:
> [this] work is a reminder that no matter how elegant or infallible mathematics may seem, it’s still a human endeavor.
Seems a little dramatic, but cool article, thanks for sharing .
> Everybody knows that this is a pandemic of the unvaccinated. But if hospitalizations and deaths were almost exclusively occurring in the unvaccinated, why would booster shots be necessary?
There is a simple example of how both these “seemingly contradictory” statements can be true.
1. Suppose that being unvaccinated increases chance of death by a factor 10. Then, it would reasonable to call the pandemic a pandemic of unvaccinated
2. Suppose that the booster decreases the chance of death by another factor of 10. Then, many vaccinated individuals may still want the booster.
The authors argument is a example of reductio ad absurdum. The author attempts to argue against the vaccine by trying to show that getting the booster leads to a logical contraction by using a poorly reasoned absurd example.
I am disappointed to see such a poorly reasoned articles on HN.
https://landing.google.com/advancedprotection/
When you login you are required to use a security key (like Yubi key) but it removes all the annoying emails and texts with codes, IP filtering, login AI, etc
When I was an undergrad taking an advanced class about probability theory, I asked my professor for help understanding the bigger picture. I could solve each of the problem sets, but I couldn’t see the bigger the picture. Why the hell are we doing this? The professor told me something like “Oh don’t worry, somethings are just impossible to fully understand the first time. Once you take a second and third class that uses these ideas the bigger picture will come together”
I have found this mindset to be incredibly true. Rather than philosophizing about the optimal way to learn to code (or anything) just:
1. read/take a class about the subject
2. use the ideas you learned
3. goto 1
While it seems inefficient, I think it can be a very natural way to learn and avoid all catch-22 situations
Whenever someone says a particular machine learning application is not going to work, and their justification is something a philosopher said, I always laugh.
I’m sure this philosopher was a great thinker with many interesting ideas about human nature, but likely they had little familiarity with convolution neural networks.
It’s unclear why the criticism in the paper does not apply to many existing “AI” technologies.
For example, clearly, video filters (like those developed by Apple, Meta, Snapchat, etc) that create a cartoon figure that mimics the movement of a users face are impossible since there is a temporal component and we cannot possibly perfectly represent every single face perfectly? Tesla self driving cars already work. Sure they might crash sometimes, but humans also crash sometimes. At this point is only a matter of time until the crash rate is deemed acceptable
I think it would be better to say
"about 3% of US workers quit their jobs in November"
For context, from Jan-2000 to Jan-2020 the average monthly percentage of workers quitting was x%"
Can anyone find a reference for x? I am having trouble.
The example graphic from the article [2] indicates that a Doner costs $11 and produces $1 profit with a labour cost of $2.50.
If the business is priced to maximize profit, this seems to imply that if the price was raised to $12, the revenue would decrease by 1/2, which is surprising.
Also, if the labour cost is 250% of the profit, then it is surprising that decreasing hours of operation to reduce labour costs would not increase profit.
Does anyone have thoughts or references explaining this?
[1] https://www.investopedia.com/terms/d/demand-curve.asp
[2] https://miro.medium.com/max/700/1*BgXwk5MMwMxqgNGAIxtplA.png
Since you are curious, here are two reasons:
1) I regularly use multiple devices, and it is convenient. For example, lately local business require online orders/reservations. If you set up an account up using your laptop, then you can immediately login to the account on your phone.
2) I trust the math behind online password managers. For example, see this white paper describing 1password https://1passwordstatic.com/files/security/1password-white-p... In particular, see pages 10-11 (two-secret key derivation 2skd) and page 18 (How Vault Items Are Secured). As long as the 2skd protocol is implemented correctly, then it should not matter if the password manager is hacked. Presumably the few dollars a month you pay for the service is used to pay someone to carefully check this code.
Edit: In theory 2) reduces the "attack surface" to just the code implementing the 2skd protocol
They are only arguing WHAT is important to teach.
High school math teachers do not have the perspective to understand what kind of math is needed for jobs in engineering, data science, etc (The fact is that a background in algebra and calculus is necessary for almost all of these jobs).
Is there data to backup this claim?
In my experience most of the top students at top universities attended private high schools. (I went to public school)
This might be enough to make the average private school student smarter than the average public school student.
It seems likely that the median public school student is smarter than the median private school student, based on the same reasoning in the tweet (but that is less encouraging)
This step of the argument assumes the conclusion. If both people answer n true false questions randomly, then with high probability (probability at least 1- 1/sqrt(n) ) one person will have answered more questions correctly.
This does not imply that either person is better than random.
There is nothing impressive about beating Wesley So according to another chess meme that became popular last year:
> "w"esley "s"o is nobody for me, just a player who are crying every single time when loosing
https://en.chessbase.com/post/cheating-controversy-at-proche...
You can see the games related to the controversy here:
> Modeling was conducted with a variety of machine learning algorithms using the scikit-learn library: K-Nearest Neighbors, Naive Bayes, Logistic Regression, and DecisonTreeClassifier, as well as XGBoost and LightGBM. Models were trained and tuned on the train and validation splits, evaluating only once on the held-out test set for our final numbers.
Direct link to ROC figure:
https://journals.plos.org/plosone/article/figure?id=10.1371/...
> Circular Harmonics are just a Fourier series
The term circular harmonics is just a nonstandard name for Fourier series.