NarcolepticFrog

6 ms·

1.
by NarcolepticFrog 7y ago
This seems to be an unfair comparison to me. The penalty for not paying the fare is not meant to somehow recuperate the lost cost - instead, the penalty is there to incentivize people to pay the regular fare, rather than skipping it. The re
2.
by NarcolepticFrog 8y ago
I don't know if I'd say it's irrelevant if it's not used by the current industry. It's always helpful to understand what is possible and how to achieve it. And beyond actual application, formalizations of fairness n
3.
by NarcolepticFrog 8y ago
That's a really interesting point - it does seem like most people immediately jump to skynet or our robot overlords whenever the topic of AI comes up. I don't have any really solid suggestions for what a good mental framework for
4.
by NarcolepticFrog 8y ago
I whole-heartedly agree. This is one of the reasons I think that education about AI and ML is important (and, on the flip side, I also think that it wouldn't hurt to educate AI and ML researchers/engineers about humanist issues so
5.
by NarcolepticFrog 8y ago
I agree with the sentiment that the discussion about AI and Machine Learning should not be entirely driven by industry. At the same time, this article seems to be a bit down on AI itself, and part of their message is that AI doesn't
6.
by NarcolepticFrog 8y ago
Why would it? Eigen-faces tend to not even look like faces (which makes sense, given that the goal is to represent a collection of real faces as linear combinations of the orthogonal eigen-faces). One possibility would be to generate random
7.
by NarcolepticFrog 10y ago
I think it depends on what you mean. I don't think that just the number of bytes of data should be used to measure the size of your dataset. For example, if I have 1TB of all 1's, this is a lot of data, but not very interesting. I
8.
by NarcolepticFrog 10y ago
What's unreasonable and offensive here is baselessly claiming that an institution is arresting its students and alumni to keep them quiet.
9.
by NarcolepticFrog 10y ago
No offense, but I don't believe you. Do you have some evidence?
10.
by NarcolepticFrog 10y ago
I feel like you must be leaving part of the story out - you don't get arrested for complaining...
11.
by NarcolepticFrog 10y ago
Could an approach like this (with caching the compiled binaries) potentially be a solution to the emacs dumping problem?
12.
by NarcolepticFrog 10y ago
I read further down the page - they use a quasi newton method to find inputs that minimize the (squared?) distance between the desired output and the output of the guessed input. I guess this can only really be expected to work well when th
13.
by NarcolepticFrog 10y ago
I'm curious about how the backwards computations work - how do you detect when the backwards computation can be easily calculated? For example, if we hash some input string and then change the value of the hash, it should be very diffi
14.
by NarcolepticFrog 10y ago
I agree that there are important differences between human and computer languages and that we should be mindful about the gap, especially when we try to automate cognitive tasks like interpreting contracts (the DAO!) , assessing beauty, dec
15.
by NarcolepticFrog 10y ago
dedicated google cardboard viewer!
16.
by NarcolepticFrog 10y ago
In general it is a misconception that real data always follows a normal distribution. It is true that if you sum many /independent/ random quantities, then the result is approximately normal (e.g., the central limit theorem and ge
17.
by NarcolepticFrog 10y ago
I agree. For both the MAD and STD, we are trying to reduce information about the "spread" of a distribution to a single number. Any such reduction must lose information, so you should pick whichever quantity is suitable for your n
18.
by NarcolepticFrog 11y ago
This is awesome! I worked on a very similar project a few months ago, which I also posted to HN: https://news.ycombinator.com/item?id=10418829 Looks like yours is quite a bit more polished :)
19.
by NarcolepticFrog 11y ago
The missing closing quotation mark was the worst.
20.
Is JavaScript Bitcoin mining a viable alternative to online advertisement?
2 points by NarcolepticFrog 11y ago|2 comments
21.
by NarcolepticFrog 11y ago
Yeah, it looks pretty similar! I guess that filter probably does something similar, except with uniformly sampled sites. Thanks!
22.
by NarcolepticFrog 11y ago
You're, right, the idea of the challenge seems to be almost identical :) Thanks for the link - their images are awesome!
23.
by NarcolepticFrog 11y ago
Heh, good point. I'll do that when I get home tonight. Thanks! (Edit: done)
24.
A fun image-processing project marginally related to my learning theory research(github.com)
36 points by NarcolepticFrog 11y ago|7 comments
25.
by NarcolepticFrog 11y ago
You're right, the results look quite similar! Thanks for pointing it out!
26.
A fun image-processing project marginally related to my learning theory research(github.com)
7 points by NarcolepticFrog 11y ago|2 comments
27.
by NarcolepticFrog 12y ago
Yes, exactly. More generally (and for the same reason), any sequence of iid random variables will form a Markov chain. For an example without the Markov property, consider the sequence of random variables X_1, X_2, ... with X_1 being either
28.
by NarcolepticFrog 12y ago
It's nice to see an explanation of Markov chains (and some techniques for predicting their behaviour) targeted at a wide audience. My only complaint is that the provided definition of a Markov chain is somewhat unclear and imprecise (a