au8er · HackerTrans

This just again shows that given enough time skill, and resources, any security is pointless if the attacker has physical access to the device.

au8er

geçen yıldiscuss

The "book" is accompanying studying material for the course Probabilistic AI at ETH Zurich. Essentially each chapter is the material covered in one lecture (3hrs).

Source: I did the course

au8er

2 yıl öncediscuss

Often times chirality is not just a fun quirk. For certain drugs, the left handed and the right handed versions can have completely different effects on the human body: https://en.wikipedia.org/wiki/Chiral_drugs#Drug_toxicity. To avoid making the wrong type of drug, a specific way of producing the chemical molecule has to be derived: https://en.wikipedia.org/wiki/Enantioselective_synthesis.

au8er

2 yıl öncediscuss

There is no independent scalar floating point unit for most modern CPUs. When scalar floating point arithmetic is needed, it is send to the SIMD unit. This pretty means that scalar and vectorised floating point operations usually have the same latency. If you do any scalar floating point operations, the CPU is just doing vectorised operations except with only 1 useful value.

au8er

2 yıl öncediscuss

If every other exchange is selling $AAPL at $100 and suddenly the top level of one exchange drops to $99, then if you just take out that order you basically gain a free dollar. Do this very fast and have pricing the product accurately and you will print tons of money.

au8er

3 yıl öncediscuss

It's the predecessor of Go1. If you buy the Pro/Edu version of Go1 then you get direct low level SDK control, though there are reverse engineering efforts that can do the same on the cheap ~5k version as well

au8er

3 yıl öncediscuss

Would it be better to do

  sparse_val = sparse[i]
  return sparse_val < n && dense[sparse_val] == i

Regardless, we are treading in undefined territory, and the behaviour is not something that can be reliably be dependent upon

au8er

3 yıl öncediscuss

For a similar simulation but using a different method, you can find one here[0]. It obviously lacks the force control, but has the viscosity tunable parameter.

[0]:https://matthias-research.github.io/pages/challenges/fluid2d...

au8er

3 yıl öncediscuss

Step 3.5, see some other interesting paper is referenced in the related work, go to step 1.

au8er

3 yıl öncediscuss

While the github contains the code, the article describing the optimisations are here: https://machinelearning.apple.com/research/neural-engine-tra....

TL;DR: execution of pytorch models on apple's neural engine and standard data-oriented optimisations (changing matrix layout, chunking to optimise temporal cache locality, and minimising redundant memory copies)

au8er

3 yıl öncediscuss

From my experience, saying "I'm thinking" actually makes it worse. Instead of thinking for a response, I internally start to panic (mainly thinking whether I have considered everything and the response is correct). After a few seconds of "thinking", I eventually give a response that is no different to my knee-jerk reaction.

I do find replying with an email to be extremely helpful. Even if the response is not correct, it does show you have put in the effort to reflect on the meeting after it has finished.

au8er

4 yıl öncediscuss

While learning, hiding details is important, but Python overdoes the abstraction. `for` loops and iterator based loops have been merged into one thing, and I am sure people write `for i in range(10)` without a good understanding of what the `range` function does. Even strings blur the line between individual characters and actual strings. The lack of clarity at these basic concepts is precisely why Python is not suitable as a first language, since it hides concepts that are prevelant in all other mainstream languages. As a result, I have often seen students develop bad coding habits and poor mental model of why their code works. Instead of learning proper programming concepts, they just know the syntactic sugar and abstractions that Python offers.

I do agree that Python provides a very small startup cost to writing code, the details are sacrificed to acheive this. This may be useful to capture interest/generate motivation and get a quick prototype/"helloworld" out, learning Python is only good for learning Python, not for programming in general.

au8er

4 yıl öncediscuss

I feel like python has never been a good introduction language. It is simply too abstract and hides too many details that are often required in other languages (Python is simply too flexible). It is good to learn if you are only scripting or calling library API, but I feel like it is a poor starting point if the goal is to transition into languages with more rigid structures such as C++.

au8er

5 yıl öncediscuss

As mentioned in the article, he aims to make each act fit in 40 pages, not the entire screenplay.

> Roth also said the 40 page limit helps him structure his screenplays.“I like it because it makes acts,” he said. “I realize if I hadn’t said it in 40 pages I’m starting to get in trouble.”

au8er

5 yıl öncediscuss

I guess just like the Panama papers, leading journalists will be killed, investigations stop being reported in mainstream media, and the world goes quiet again with nothing changing.

au8er

5 yıl öncediscuss

You cannot use unit tests for everything, since the number of unit tests required is basically infinity. Logically, unit tests can only prove an error exists and cannot show the program is correct. What they most likely did was mathematically showed that all possible edge cases (potentially inifinitely many) are all considered.

au8er

5 yıl öncediscuss

The author, Onur Mutlu, is at ETHZ now and he livestreams all his lecture on youtube. He mainly covers computer architectures.

https://www.youtube.com/channel/UCIwQ8uOeRFgOEvBLYc3kc3g

au8er

5 yıl öncediscuss

This is comparison between US and (China, India, and Russia). The original poster was comparing between US and Europe, which this paper does not cover at all.

For reference, the title of the paper is "Computer science skills across China, India, Russia, and the United States"

au8er

6 yıl öncediscuss

From what I recall in my neuroscience class, this is likely to be caused by damage to the visual cortex, which is used to consciously process visual information, but not the superior colliculi, which is used to unconsciously process peripheral vision. This is caused by damage to one of the visual processing regions but not others.

From the article, it appears something similar has happened. Corticobasal syndrome potentially killed off some brain cells but not others, leading to damage at some areas but not others.

comments