HNHacker News
TopNewBestAskShowJobs

mziel

358 karma · joined March 1, 2014

submissionscomments
mziel··on Jobs Involving Routine Tasks Aren't Growing
Could you post a link to this talk?
mziel··on Are coders worth it? (2013)
Agreed. Our hourly wage does not equal the amount of hard work during that hour. It's also a (amortized) function of all the years you've spend in or outside of school perfecting your skills.
mziel··on Digitally weary users switch to ‘dumb’ phones
Not a FB user, but actually it can't be uninstalled.

I had an Android phone with no FB (I probably deleted it after I got the phone), and after the system upgrade it got installed without me agreeing to it. And I can't seem to undo it.

mziel··on 2016 Spark Summit East Keynote
Last Spark Summit the videos were up on Youtube 1-2h after each talk. Anybody knows where to find the ones from this summit?
mziel··on R, the master troll of statistical languages (2012)
If you want to go over data-frame in parallel just replace call to lapply() with mclapply().

I agree about 3 OO systems, but let's not mix things up here. Casual user (who doesn't know sapply) doesn't interact with that.

mziel··on R, the master troll of statistical languages (2012)
YMMV, but I think this is pretty clear:

"lapply returns a list of the same length as X, each element of which is the result of applying FUN to the corresponding element of X" == lapply is a map() construct that takes a list and a function

"sapply is a user-friendly version and wrapper of lapply by default returning a vector, matrix or, if simplify = "array", an array if appropriate, by applying simplify2array(). sapply(x, f, simplify = FALSE, USE.NAMES = FALSE) is the same as lapply(x, f)." == "sapply(x, f, simplify = FALSE, USE.NAMES = FALSE) is the same as lapply(x, f)"

mziel··on R, the master troll of statistical languages (2012)
For R at least some of it is due to R's flexibility

x$name syntax stems from data frames being really lists in disguise x[["name"]] ditto, plus because it's useful to access by string (see reflection in other languages)

x[,"name"] and x[,1] because we can also apply the matrix syntax to data frames

mziel··on R, the master troll of statistical languages (2012)
I think it's also due the fact that installing R package is just one command away, install.package("abc") and library(abc) and you're done. It a blessing, but encourages loading swaths of libraries.
mziel··on R, the master troll of statistical languages (2012)
Factor is one of the worst thing of R-world. I don't recall ever needing factors, yet they creep in with many functions (read.csv, cut).

Btw there's a nice readr package (from Hadleyverse) that has a read_csv method that does away with factors by default.

mziel··on R, the master troll of statistical languages (2012)
The problem is people using R without trying to learn about the language itself, just assuming it works like their favourite language.

For example complaining that R is slow and then writing iterative solution instead of using vectorization. When I saw the example the author gave my first thought was "sapply/lapply". Lapply is essential to the R use, and is being taught early on in every book/course on R I've ever saw.

"In 2012, I’m the kind of person who uses apply() a dozen times a day, and is vaguely aware that R has a million related built-in functions like sapply(), tapply(), lapply(), and vapply(), yet still has absolutely no idea what all of those actually do. "

mziel··on PyThor- Python meets R
And lists + dataframes as first class citizens. Everything (including dataframes) is pretty much a list. Most R packages nowadays have dataframes in the centre of implementation.

Compare it to other languages, lots of code is needed to convert the data from one format to another, because there's no underlying data structure (Python is much better than the rest with regards to that, but still not as good as R).

mziel··on Announcing Spark 1.6
That sounds very interesting. Is there any ticket, where I can follow the progress on the batch columnar format you mentioned?

Btw, I was critical about the issue above, but I do love Spark, using it on a daily basis. :)

mziel··on Announcing Spark 1.6
New statistics/machine learning algorithms are always welcome, but the big plus for productionizing is ML pipelines persistence.

That said, during Spark Summit Databricks guys themselves were most excited about Dataset API. Looking forward to giving it a try.

mziel··on Announcing Spark 1.6
To be honest PySpark and SparkR are always going to be 2nd category citizens (because of the serialization/pickling between the two environments). Databricks shows nice graphs, saying they are equivalent for DataFrames, however those count only for built-in functions that basically translate code into execution plan for Catalyst. For anything bespoke (UDFs, custom Transformers/Estimators) you're better off using Scala.
mziel··on Uber’s No-Holds-Barred Expansion Strategy Fizzles in Germany
Just.

What about the coordinating those? What about review system for drivers?

mziel··on Switzerland to vote on banning banks from creating money
Not the OP, but there's no such thing as proper experimental setup in economics (because there can't be). That's why using economic/econometric modelling for causal inference is the holy grail of the field.

Some approaches are:

- structural models (google Pearl)

- instrumental variables (google Angrist)

mziel··on How MTV's “16 and Pregnant” May Have Reduced Teen Pregnancy
I think this is happening to a degree. I know of a guy, whose younger brother started reading up on physics after Big Hero 6.

Also didn't older tech-related movies on hackers inspire some of us?

mziel··on How to trick a neural network into thinking a panda is a vulture
For the lazy:

http://research.microsoft.com/pubs/255952/ICCV15_DeepNDF_mai...

From abstract:

"(...) we introduce a stochastic and differentiable decision tree model, which steers the representation learning usually conducted in the initial layers of a (deep) convolutional network. (...)"

mziel··on Ask HN: Books you read in 2015?
Not-technical

"The World's Toughest Endurance Challenges" - Very nicely illustrated

"Thing Explainer: Complicated Stuff in Simple Words" - Extremely enjoyable. People here certainly are aware of xkcd and it's type of humor.

Technical

"Functional programming in Scala" - good read for Scala beginners, especially after "Scala for the Impatient"

"Learning Spark" - the best book on Spark so far

"Assessing and Improving Prediction and Classification" - couple of interesting ideas for ML and Data Science

"Neural Network Design" - Quite basic, but I like the flow and introducing mathematical concepts just before they're needed.

mziel··on The Hadoop Ecosystem Table
Nice list, but Spark is treated superficially. Also extremely out-dated (Shark, Bagel).

SparkSQL should be in the SQL-on-Hadoop section. MLlib+ML should be in Machine Learning section. If we include Storm and Giraph, we should include SparkStreaming and GraphX.

mziel··on The strange non-transitivity of random variables (2014)
It's not. Let's assume that ">" means probability of winning over 0.5

Federer > Djokovic (well not lately)

Djokovic > Nadal

Nadal > Federer

Everything depends on the draw. Same with the initial choice of dice.

mziel··on An Open Letter to California Air Resources Board on VW Cheating Scandal
Depends on your settings. (right click -> options)

I have uBlock (additional lists) and Ghostery (block all) and don't see that.

mziel··on Should AI Be Open?
See other comments. Signed admission doesn't mean anything without 3rd party verification.

I've never seen a published log. Link?

mziel··on Bayesian Logic Programming
I don't think they understand confidence intervals, or at least they think they do but they get it wrong. It's the same as misunderstanding p-value.

Confidence interval of 95% means that the estimator produces an interval that contains true parameter with probability 95%. It's not equivalent to the credible interval.

https://stats.stackexchange.com/questions/2272/whats-the-dif...

mziel··on Should AI Be Open?
Yes, I'm aware of that. But without 3rd party we're basically trusting the participants that it didn't happen.

E.g. "I'll give you $500, but you also have too sign an NDA, so that people don't know we cheated."

I don't want to imply that they cheated, just want to reiterate my original argument that the lack of transparency makes the experiment effectively invalid. Think Tesla and his claims about World Wireless system.

mziel··on Should AI Be Open?
And yet Big Yud refuses to publish the conversation. I know his arguments (unknown unknowns) but this is a very un-scientific approach and frankly why should we believe that what he said happened really happened?
mziel··on The Importance of Reading as a Software Engineer
Logic tells us those two statements are equivalent.

p = "reading books"

q = "success"

p => q is equivalent to ~q => ~p

mziel··on Bullshit Startup Experiences
Not trying to start a flame, but what difference does it make that they're male or female? The difference is as you said "some important school".
mziel··on Comparative Study of Caffe, Neon, Theano, and Torch for Deep Learning
This very much. Understanding the underlying concepts is crucial. Many people use out-of-the-box tools just plainly wrong, because they didn't bother with fundamentals. I know a guy who reinvented Naive Bayes and Laplace smoothing. Kudos for him, but he wasted about 3-4 months of company's time on it.

That said, I would start with Python based frameworks:

- Caffe is more for production, and Theano and family for experimentation.

- You can play with Theano or Tensorflow using Keras which abstracts them

- Python scientific stack makes it easier to visualize the results later

mziel··on Books Are Getting Longer
You misunderstood me. I was not criticising the submitter, kudos for the change. I was criticising the clickbaity title of the article.
← PreviousPage 2 of 4Next →