ganfortran

11 ms·

61.
by ganfortran 10y ago
> for inference True, but for training, in near future, it is still going to be GPU, probably Nvidia GPU only, not too much Intel could do. > requires a fair amount of domain expertise to get good performance out of GPUs Rightfully so
62.
by ganfortran 10y ago
University students optimal learning time of the day varies person by person, common sense indicates.
63.
by ganfortran 10y ago
Wechat embedded a browser in itself. In that since it is more like Chrome Web Store.
64.
by ganfortran 10y ago
Wechat controls social connection. It is like you cannot compete with Facebook on social as well here. So, you have to cooperate with them.
65.
by ganfortran 10y ago
So...television, is what your saying. Funny that we went all the back to the starting point. I do agree programmatic ads is a loop of shithole, driving both content and ads into a mutual destruction cycle. Brands are hurting, malware ads ar
66.
by ganfortran 10y ago
Where is GM's product?
67.
by ganfortran 10y ago
I guess computer is much more deterministic than what is required for ML to be useful. ML, in a very inaccurate way, can be seen as: 1.We have observations and conclusions. 2.We don't know exact those observations leads to the conclusi
68.
by ganfortran 10y ago
20 hours sounds right. I don't think I could gain more actual productivity forced by someone even myself to work long hours. Often time, when it feels right, the code just spills itself like charm, and after that my brain just turned s
69.
by ganfortran 10y ago
It is more about ideology. The 3.x is catching up, but still not fast enough as the community wants it to. But 2.7 is here for a reason, and shaming people that used 2.7 to push 3.x adoption, in my opinion is a bad practice.
70.
by ganfortran 10y ago
Which is fine. It is DL library, not for string processing.
71.
by ganfortran 10y ago
Which is already the case like for 10 years. That is why no Startup ever challenge Google for their Search. It is basically the Matthew Effect but on Web Scale.
72.
by ganfortran 10y ago
If human fuck up Earth, I don't think Mars will survive either.
73.
by ganfortran 10y ago
> prices of land a ludicrous Well, trip to Mars won't be cheap, and by the point it is inhabitable for average people, it would have land owner maybe government as well, at that point.
74.
by ganfortran 10y ago
Yeah, that is my question. Why living on Mars, where you don't even have drinking water available? Human population in a lot countries, such as Japan/China/Korea is going to shrink pretty soon. In US, without immigrants, the
75.
by ganfortran 10y ago
LOL, Russian Government apparently don't understand how Internet really works.
76.
by ganfortran 10y ago
Tax the robot
77.
by ganfortran 10y ago
And why is this News. They don't have a product to sell other than a toy demo. No chatbot company can offer an easy to understand, measurable interface for customers to customize towards their own usage case, with expected outcome. You
78.
by ganfortran 10y ago
Math needs to fulfill a purpose in order to succeed in INDUSTRY. A image recognition is first and foremost about recognizing an image, the math behind it is mystical, but the accuracy is measurable. If you want your knowledge to be hold as
79.
by ganfortran 10y ago
Yes we do. In terms of spending time mathurbating what is fundamentally limited and incapable, I suggest we spend more time play with the shiny new toys that do fullfill its very own promises.
80.
by ganfortran 10y ago
Neural Architecture Search with Reinforcement Learning https://arxiv.org/abs/1611.01578 Too bad, it is already there :P
81.
by ganfortran 10y ago
Too bad that Math heavy approaches just cannot compete.
82.
by ganfortran 10y ago
What does this imply?
83.
by ganfortran 10y ago
1.What is your estimate then? Hundreds Dollars? Thousands? 10 Thousands? Surely it might fall into the later two category, since paying a professional speaker to sit and work for 26.4 hours is already over 1000 dollars, if you assume 40$ pe
84.
by ganfortran 10y ago
I see your points. But it is related. Even if u get what you think the paper describes, it is hard to know whether you did it right or not, because you cannot replicate the result easily. This happens in a lot of CV papers already, where pe
85.
by ganfortran 10y ago
> How difficult Very difficult: 1.Machine learning is pretty data dependent, and make those datasets are very expensive. Google is not likely to give them away for free, because it is their competitive advantage. 2.The infrastructure to
86.
by ganfortran 10y ago
No benchmark? I don't really think this hold against Redshift.
87.
by ganfortran 10y ago
Do they train with backpropagation efficiently?
88.
by ganfortran 10y ago
> On the other hand, maybe people aren't actually that bothered about which particular ride company they use Which is indeed the case here. Why people would prefer one to another? Cheaper, or more drivers or both. Brand? Simply don&
89.
by ganfortran 10y ago
Well, it all starts with the work harassment blog, which is entirely personal and nothing to do with media. The Waymo lawsuit too. Uber did those shitty stuff itself. Media did exactly what it supposed to do, to spread the information to a
90.
by ganfortran 10y ago
> If you want a popular AI topic which does require a lot of inference, think of a general-purpose chatbot. General-purpose chatbot, almost equal to solve AGI. If that is how you define winter...where is summer!?

More ›