iskander

41 ms·

391.
by iskander 12y ago
The FPÖ seem like some old-school xenophobes and antisemites. Campaigning under creepy slogans like "more strength for our Viennese blood" and publishing cartoons with hook-nosed banker Jews. In what way are they getting more cent
392.
by iskander 12y ago
>This is a war memorial. How might it feel to those who have lost loved ones in one of this country's innumerable wars, to have the monument to their loss, their sacrifice, appropriated for someone else's speech? This is the to
393.
by iskander 12y ago
With other people! Impenetrably obscure papers really unravel in a journal club. You can understand the main thrust of a paper much faster if you find a social setting for talking about it.
394.
by iskander 12y ago
Spark has a much nicer API than Hadoop and theoretically can give you significant speedups on iterative in-memory work loads. On the other hand, I've had a terrible experience with the stability and debuggability of previous versions o
395.
Workshop on array-oriented programming languages(sable.mcgill.ca)
3 points by iskander 12y ago|0 comments
396.
by iskander 12y ago
>That they can improve drug discovery? ICML is not a molecular biology journal and can't judge this paper on those merits (and probably neither can most people on news.yc). From a machine learning perspective, though, there isn&#x27
397.
by iskander 12y ago
Rich Caruana's original paper on multitask learning (from the 90s) talks about this very kind of multi-headed neural networks and it doesn't appear that this Google paper has done anything new with the algorithm. So, what's a
398.
by iskander 12y ago
>no differences (or they are extremely small) ...other than dramatically altered patterns of gene expression, alternative slicing, epigenetic marks, and irreversible differences in phenotype. Genetics is a small part of the picture and m
399.
Exploring the Genome with Ensembl and Python(hammerlab.org)
1 points by iskander 12y ago|0 comments
400.
by iskander 12y ago
>if people who were, for example Christian, that decided there was no God took the good parts of the religion (peace, forgiveness, some of the rituals etc.) and created a branch of it. Maybe it's already been done? This is a decent
401.
Guacamole: distributed genomics with Scala and Spark(hammerlab.org)
3 points by iskander 12y ago|0 comments
402.
by iskander 12y ago
1pb / 100gb / genome = 10,000 genome sequences. And that's just raw data from one platform. If you're interested in e.g. splicing diversity you would want to do long read RNA sequencing. Leaving room for intermediate res
403.
by iskander 12y ago
Genetic sequence data, mostly from cancer. (Tools are terrible, data sizes up to hundreds of gigabytes per patient)
404.
by iskander 12y ago
I wonder if it's possible to remove the need for post-processing of the LSTM's output by integrating transcription into the neural network model directly.
405.
by iskander 12y ago
Sometimes "hypothesis" persists well past the guessing phase (e.g. "The Hygiene Hypothesis" is pretty well supported, even if it's not a complete explanation of autoimmunity and allergies)
406.
by iskander 12y ago
This would be more interesting if they had teased apart more informative axes of belief instead of categorizing people along the uninformative liberal vs. conservative spectrum of US politics. Do social psychologists believe that the class
407.
by iskander 12y ago
This bit stands out to me as prescient: >The idea, though, that improved technology is going to solve the problems that the technology has caused seems to me to be a bit quixotic. I understand that there’s a certain amount of hope about
408.
by iskander 12y ago
As a Jew whose family tree mostly stops in 1941, I think understanding the psychology of Nazi (both the rank-and-file and their leadership) is extremely important. The Nazi phenomenon is not an anomaly, it arose from psychological biases
409.
by iskander 12y ago
Racism is bundled up with a bunch of cognitive biases and low-level instincts, which makes it dangerous. People have a hard time distinguishing between a particular fact, their significantly broadened interpretation of that fact and its con
410.
by iskander 12y ago
I strongly recommend connecting profile pages to github and/or technical blogs.
411.
by iskander 12y ago
I just drove from Portland to San Francisco, and many of the bridges along Highway 101 looked similar to these photos. Also, in the small towns I stopped in, the drought was a common topic of conversation. Locals said they had never seen th
412.
by iskander 12y ago
I was an intern on the Google Health team and witnessed many problems unrelated to external barriers. 1) The original codebase was a nightmarish mess of Java/GWT code that did very little (my first real-life encounter with a FactoryFac
413.
by iskander 12y ago
>If I were a demon from Hell, charged by my infernal masters with increasing rape as much as possible, I literally could not think of a better strategy than talking about rape culture all the time. I don't understand the proposed al
414.
by iskander 12y ago
Thanks for posting this. I'm starting to get a feel for when Spark is usable-- you need an underlying indexed data store which lets you fetch small subsets of your data into RDDs (or, your data can be tiny to begin with). We've be
415.
by iskander 12y ago
> that Spark performance has been great when I've used it. Can you say more about your use case? What sort of data did you start with? What did you do with it? How large was the cluster you were running on?
416.
by iskander 12y ago
I'd love to learn more about how they're using Spark, are there are any blog posts or tech talks floating around?
417.
by iskander 12y ago
Well, the RDD is initially partitioned using a RangePartitioner over a dense key space of Longs. Each element is then expanded ~100x (each object is significantly smaller than the original value). So the total memory footprint and skew of t
418.
by iskander 12y ago
> For sub-second adhoc queries we go against bitmap indexes. Did you implement the index yourself? How many events survive the initial filtering?
419.
by iskander 12y ago
> It is a rock solid platform. Our hosted ops loves it because it just works. I'm really confused by how different our experiences have been. Above you said you wrote your own shuffle implementation, presumably that was prompted by
420.
by iskander 12y ago
So the central piece of data is something like a 10 million element RDD of (UserId, (MovieId, Rating))? If so, it sounds like that data would fit into a single in-memory sparse array, how does Spark's performance compare with a local i

More ›