argonaut

88 ms·

451.
by argonaut 10y ago
> Personal anecdotes > observed facts These are not the same thing.
452.
by argonaut 10y ago
Except nobody is claiming all runners need a high carb diet. If anyone actually claimed that it was that most runners would benefit from a high carb diet (I have no idea if anyone actually makes this claim scientifically).
453.
by argonaut 10y ago
Time for a funny quote: "Power laws... have been termed 'the signature of human activity'... They are certainly the product of one particular kind of human activity: looking for power laws." http://people.seas
454.
by argonaut 10y ago
There's a certain irony in calling something a myth based on your personal anecdote.
455.
by argonaut 10y ago
ISLR was written after ESL to be an easier to read version. I guarantee you will have trouble if you try to teach yourself ML with ESL.
456.
by argonaut 10y ago
Not sure why this is downvoted. ISLR is widely considered the "easier to read" / more pedagogical version of ESL. It's still mathematical but is closer to what the poster wants - a textbook that teaches them.
457.
by argonaut 10y ago
I hear people recommend textbooks a lot, and I honestly don't know why. Very few researchers I know learned machine learning through reading a textbook on their own. Furthermore, the first two textbooks are closer to reference books ra
458.
by argonaut 10y ago
No, the computer can't. The computer was itself trained/programmed by humans.
459.
by argonaut 10y ago
Yes, if they meet certain reasonable conditions (e.g. they're identified separately on the bill).
460.
by argonaut 10y ago
The main reason we see all these relatively information-less blog posts / tutorials is because most people on HN are afraid of the math. If you're not afraid of the math (or have a relatively strong math background), then honest
461.
by argonaut 10y ago
According to the article, tips must go only to front house staff by NY state law.
462.
by argonaut 10y ago
As the article says, waiters at upscale restaurants are worse off in this system, because the system is now less unequal in their favor. If this became a trend at lower end restaurants, waiters that don't make much in tip would mak
463.
by argonaut 10y ago
No, that is a ridiculous comparison. Then you should start counting the energy required to construct the Google server farm, the energy of all the computers used by all the engineers who built the farm while they were in university, on and
464.
by argonaut 10y ago
Except these firms don't use LSTMs at their core. They do tons of feature engineering combined with statistical models and very simple machine learning models (think linear/logistic regression). Some firms will use more complicate
465.
by argonaut 10y ago
According to Google they used 280 GPUs and 1,920 CPUs. I'm sure they used custom chips in addition.
466.
by argonaut 10y ago
Firstly, there are different approaches. Google and Uber seem to have a similar LIDAR + map approach. Tesla and Mobileeye have a camera focused approach. The CEO of Mobileeye (who is an ex-machine learning professor) gave a very good, mildl
467.
by argonaut 10y ago
There is a big open research problem in the state of the art in visual SLAM (visual SLAM = SLAM from cameras), it doesn't work when the environment is moving (!!!). Visual slam is still linear-algebra/geometric/keyframe bas
468.
by argonaut 10y ago
Exactly. After going to Berkeley, there are 10+ people at top tech companies who think I'm smart and would be more than happy to refer me if I needed a job.
469.
by argonaut 10y ago
> Humans only have one sensor: a rotatable stereo camera. So at least in theory the number of sensors seems not the most important element ;-) I hear this comment a lot when defending Tesla's choices, and it's a red herring. Th
470.
by argonaut 10y ago
Many of the top students at Berkeley, Stanford, MIT don't get standard SWE roles. They go to quant finance and make 300k straight out of undergrad. They go to PhD programs. They start a startup with YC. They do Google APM or something
471.
by argonaut 10y ago
I was referring to Berkeley, MIT, and Stanford.
472.
Whence your reward function?(nlpers.blogspot.com)
35 points by argonaut 10y ago|5 comments
473.
by argonaut 10y ago
If you had enough data, I imagine you could just re-learn the parameters on the new dataset (and also finetune the network)?
474.
by argonaut 10y ago
I went to Berkeley for CS and knew people that went to MIT and Stanford, and I don't agree. You're right the curriculum is the same. Everything else is different. Just one example: lots of undergrads do research. They get to wor
475.
by argonaut 10y ago
If you think those two anecdotes are the same, good luck!
476.
by argonaut 10y ago
To be fair, many machine learning algorithms (especially deep learning) are very parallelizable, both when training and predicting.
477.
by argonaut 10y ago
"If there's a 1% chance that Pakistani scientists are helping al-Qaeda build or develop a nuclear weapon, we have to treat it as a certainty in terms of our response. It's not about our analysis ... It's about our respon
478.
by argonaut 10y ago
1. Just because something has been published as a paper , does not mean it is applicable, says something interesting (even if theoretical), or even that it's actually correct (it just can't be obviously wrong). 2. Setting aside
479.
by argonaut 10y ago
You are confusing narrow AI for AGI. None of those things have proved anything practical about what an actually achievable AGI would look like, rather than some theoretical construct that is provably incomputable.
480.
by argonaut 10y ago
Not about AGI. We understand enough to formulate questions about narrow "AI," certainly, since that already exists in a narrow sense.

More ›