Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
tcgv
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
42 ms
·
421.
▲
by
tcgv
7y ago
Hi, I'm the author, thanks for the feedback! > The ad hoc method of parameter estimation used here needs some explaining I tried to imitate a "least squares regression" actually. I confess I wasn't sure if this was th
422.
▲
Stock Option Pricing Inference
(thomasvilhena.com)
103 points
by
tcgv
7y ago
|
16 comments
423.
▲
by
tcgv
7y ago
There are several films that are probably streamed throughout the world during christmas. One that I used to watch every year on cable TV here in Brazil when I was a child (you can guess I'm not that old) is "Jingle all the Way&qu
424.
▲
by
tcgv
7y ago
From the linked article's context: > 3. The classical computational theory of mind > The label classical computational theory of mind (which we will abbreviate as CCTM) is now fairly standard. > Turing computation is determini
425.
▲
by
tcgv
7y ago
True random number generation can't be done by a deterministic computer, and it appears that human brains can, although not conclusive yet, and the precise mechanism is unclear: - https://www.ncbi.nlm.nih.gov/pubmed
426.
▲
by
tcgv
7y ago
It took me a while to recognize that wooden chair. At first I tought it was a wooden hammer on top of some squared background, but then I realized it was a chair seen from above. For me, I actually imagine these objects moving/rotating
427.
▲
by
tcgv
7y ago
Pictures of fully grown silicon crystal ingots always impresses me: - https://ars.els-cdn.com/content/image/3-s2.0-B97801280358180... That narrow section at the top is where the crystal started growing from, and h
428.
▲
by
tcgv
7y ago
Yes it is! But incredibly slow and resource intensive. A few days ago I shared a blog post here on HN about a simple quantum computing simulador I built as a weekend project, for learning purposes: - https://thomasvilhena.com
429.
▲
Quantum Computing, for Programmers
(thomasvilhena.com)
2 points
by
tcgv
7y ago
|
0 comments
430.
▲
by
tcgv
7y ago
In an interview I watched on youtube Jim Simons says their trades provide liquidity and efficiency to financial markets, as to say they are effectively contributing to the economy, and not just earning money.
431.
▲
System Design Coherence
(thomasvilhena.com)
2 points
by
tcgv
7y ago
|
0 comments
432.
▲
Associations of sleep duration with HDL and LDL levels (2008)
(ncbi.nlm.nih.gov)
1 points
by
tcgv
7y ago
|
0 comments
433.
▲
by
tcgv
7y ago
One of the first recursive algorithms I learned was the solver for the Tower of Hanoi problem. I was stunned about its simplicity, solving a problem most people couldn't with just a few lines of code: - https://en.wikipedia.
434.
▲
Why is it hard to name classes?
(thomasvilhena.com)
1 points
by
tcgv
7y ago
|
0 comments
435.
▲
The Short Transaction Trap
(thomasvilhena.com)
1 points
by
tcgv
7y ago
|
0 comments
436.
▲
Index-Free Adjacency
(thomasvilhena.com)
1 points
by
tcgv
7y ago
|
0 comments
437.
▲
by
tcgv
7y ago
This subject has been treated so irresponsibly these past few days that not only people are making inaccurate statements such as "the Amazon provides 20% of our oxygen", but also high profile people such as Emmanuel Macron (France
438.
▲
Business Logic
(thomasvilhena.com)
2 points
by
tcgv
7y ago
|
0 comments
439.
▲
Complexity Classes of Problems
(thomasvilhena.com)
5 points
by
tcgv
7y ago
|
0 comments
440.
▲
by
tcgv
7y ago
> view ourselves in the best possible light, and to view other people as not as good Remembered me of this cognitive bias: "Overestimating Our Own Importance" [1] [1] "Psychology of Intelligence Analysis", Chapter 11.
441.
▲
by
tcgv
7y ago
> Empress trees (...) absorb about 103 tons of carbon a year per acre. In 2016, U.S. GHG emissions were 20.15 metric tons CO2-equivalent per person[1]. So basically each american would have to cultivate about 0.196 acre of empress trees
442.
▲
A Successful Deployment Model
(thomasvilhena.com)
3 points
by
tcgv
7y ago
|
0 comments
443.
▲
Using Queues to Offload Web API
(thomasvilhena.com)
2 points
by
tcgv
7y ago
|
0 comments
444.
▲
by
tcgv
7y ago
Very interesting. I was trying to think on how to refute premise 2 , and came up with: - A real catastrophe is worse than an imaginary catastrophe Then not all real things are better than imaginary things - An imaginary outcome in which my
445.
▲
Integrating Third-Party Modules
(thomasvilhena.com)
2 points
by
tcgv
7y ago
|
1 comments
446.
▲
by
tcgv
7y ago
> That's right. It's pretty amazing how much is based off of that assumption which has no realistic basis. I guess it's a "convenient" assumption. Beliefs and speculations also have no realistic basis. We can
447.
▲
by
tcgv
7y ago
Great that he found out what was making his experience on HN bad and was able to change that and contribute to others that feel the same. I personally enjoy browsing HN's comments section to read opinions conflicting to my own's,
448.
▲
by
tcgv
7y ago
That's awesome! I went to Cairo in 2016 and only got to see the Bent Pyramid at a distance from the Step Pyramid site, since I didn't have much time there. Even from afar it's an amazing sight. I'm longing to go back the
449.
▲
by
tcgv
7y ago
Sure, but only a handful of people really employ advanced calculus for building ML, while the vast majority of Data Scientist only use the ML algorithms implementations as a black box, without diving into its inner workings.
450.
▲
by
tcgv
7y ago
> I’m not sure about Brazil In Brazil most federal inspection agencies are under funded and short on staff, limiting their operations. Offering bribe to inspection agents is also very common but risky since the government has been pushin
More ›