sherjilozair

10 ms·

61.
by sherjilozair 12y ago
Is there any such collection of advanced Python patterns, aimed at Python programmers with more than 2-3 years of experience?
62.
by sherjilozair 12y ago
Because it's not important. What is more important is that the First Science Sequence was finished. In a detailed report, you may find that the battery didn't last as much as they thought it did (perhaps because of the double land
63.
Why I switched from Pylearn2/Theano to Torch7(nikopia.com)
2 points by sherjilozair 12y ago|0 comments
64.
by sherjilozair 12y ago
This is a reply to multiple sibling comments. There is actually recent work which shows that deep learning methods can also work WITHOUT any reverse signals: http://s.yosinski.com/dan_cownden_presentation.pdf
65.
by sherjilozair 12y ago
None of the parties mentioned actually deny the above equivalence. The reason backprop is a popular idea in deep learning is because people started developing continuous models, where the output (and the error) was a continuous and differen
66.
by sherjilozair 12y ago
If you're looking for programming tutorials on deep learning with some theory on the side, I found these perfect: http://www.deeplearning.net/tutorial/
67.
by sherjilozair 12y ago
Also, 3. Being different (hipster) makes the audience remember the video, and hence the product, which is the purpose of the video.
68.
Do Deep Nets Really Need to be Deep? [pdf](arxiv.org)
126 points by sherjilozair 12y ago|29 comments
69.
This Is Why You Should Never Ever Steal a Mac(thiefinkualalumpur.tumblr.com)
2 points by sherjilozair 12y ago|0 comments
70.
by sherjilozair 12y ago
This is exactly how Windows' device drivers work. What you are asking for is the same functionality be extended to userland programs. The primary problem for this is proof generation. Programming paradigms where proofs are easy to gene
71.
by sherjilozair 12y ago
A very useful addition to markdown would be the ability to put anchor links, and "open in new tab" links. Both of these are not link defaults, but are perfect examples of common cases that should work nicely. Markdown is often use
72.
by sherjilozair 12y ago
Why are close-world assumptions bad for all cases? There might be cases limited to a domain where the computer does not need to know about the world too much. Consider, the language of geometry problems given in high school math books. Sure
73.
by sherjilozair 12y ago
move left.
74.
by sherjilozair 12y ago
Could you substantiate this with any references? I thought hard about this, and I find no reason to believe that equally-spaced points would give a biased estimator.
75.
by sherjilozair 12y ago
We don't necessarily have to use randomness. We can take equally-spaced points in the quarter as well, rather than sampling. This is called pseudo-Monte Carlo.
76.
Don’t Waste Your 20s at Google or McKinsey(linkedin.com)
8 points by sherjilozair 12y ago|1 comments
77.
by sherjilozair 12y ago
Any links on relevant material? I'm okay for delving into current research topic.
78.
by sherjilozair 12y ago
I've done something similar with Sphinx, which worked quite well too. However, I feel the real challenge to productify such a technology would be to make this system recognise utterances spoken casually, over loud music, over friends t
79.
by sherjilozair 12y ago
Google. https://github.com/rcorcs/NatI/blob/master/gapi.py#L26
80.
by sherjilozair 12y ago
None of these are what I'm looking for. These do have programming exercises. However, I'm looking for a very specific type of exercise: a problem is given, and an incomplete solution is also given, and the student is asked to fill
81.
Ask HN: I can't remember this website
1 points by sherjilozair 12y ago|4 comments
82.
The Flaw Lurking In Every Deep Neural Net(slashdot.org)
2 points by sherjilozair 12y ago|0 comments
83.
Intriguing Properties of Neural Nets [pdf](cs.nyu.edu)
8 points by sherjilozair 12y ago|0 comments
84.
Devil's 2048(rudradevbasak.github.io)
2 points by sherjilozair 13y ago|0 comments
85.
Why Google Doesn’t Have a Research Lab(technologyreview.com)
68 points by sherjilozair 13y ago|42 comments
86.
by sherjilozair 13y ago
simulated annealing. ;)
87.
by sherjilozair 13y ago
If you can solve something easily, then you can also verify its solutions easily. It is a typo. It should say "but potentially extremely difficult for computers to solve."
88.
by sherjilozair 13y ago
Do you consider foreign applications? Asked this a few months back too.
89.
Levitating objects using sound waves(mashable.com)
2 points by sherjilozair 13y ago|0 comments
90.
by sherjilozair 13y ago
You're just making a fool of yourself. This work is absurd, ridiculous and legit at the same time.

More ›