instanceof me @instanceofme - Tumblr Blog

The following is a true story. Or maybe it’s just based on a true story. Perhaps it’s not true at all.

A little cautionary tale on how easy could be to propagate malicious code to websites.

Lucky for me, we live in an age where people install npm packages like they’re popping pain killers.

David Gilbertson then debunks common beliefs about how one would be supposedly safe or able to detect this.

On the challenges of teaching non-readers to read:

Here is the book’s opening sentence:

In the old brown house on the corner, a mile from the middle of the city, we ate bacon for breakfast every morning of our lives.

If you are reading this essay, you’re a reader. You probably know this sentence, and if you don’t, you are comfortable with interpreting it. You can hear a character beginning to form: its romantic, optimistic, nostalgic voice; a voice yearning for simplicity; probably, in its deliberate imitation of a child’s singsong, the voice of a woman, a mother. You know it might take a few pages to learn just who this woman is. You’re skilled in this sort of patience.

But if you have never read anything more difficult than a Harry Potter book, how are you meant to proceed?

Well, there is only one way to go on, as I tell students – and that is to go on. This is the first and greatest difficulty they face. There’s no reason for them to continue reading. There is so much else to read that is shorter, and not just aimed at them, but, in the case of their Facebook feed, tuned to their experience. Marketed to them. Why would they bother reading something that was neither for them nor about them?

#education #book

For instance:

The single target attack here changes the label of a backdoored stop sign to a speed-limit sign (such that an autonomous vehicle may not automatically stop at the sign anymore). Once again, the average accuracy achieved on clean images stays high, and the BadNets classify more than 90% of stop signs as speed-limit signs.

See also the paper

#typography

Our research started with a 96-megabyte surprise:

b97bb000-b97dc000 rw-p 00000000 00:00 0 [heap] bf7c6000-bf806000 rw-p 00000000 00:00 0 [stack]

and a 12-year-old question: "If the heap grows up, and the stack grows down, what happens when they clash? Is it exploitable? How?"

#security #linux

A smart, open source, SASS based design agnostic framework built for scale.

iotaCSS is a design-agnostic CSS framework:

OOCSS

SASS-based

BEM compliant

ITCSS-inspired

#css #sass #itcss

Cello is a library that brings higher level programming to C.

By acting as a modern, powerful runtime system Cello makes many things easy that were previously impractical or awkward in C such as:

Generic Data Structures

Polymorphic Functions

Interfaces / Type Classes

Constructors / Destructors

Optional Garbage Collection

Exceptions

Reflection

And because Cello works seamlessly alongside standard C you get all the other benefits such as great performance, powerful tooling, and extensive libraries.

#lib #c

Python & R forecasting tool

Today Facebook is open sourcing Prophet, a forecasting tool available in Python and R. Forecasting is a data science task that is central to many activities within an organization. For instance, large organizations like Facebook must engage in capacity planning to efficiently allocate scarce resources and goal setting in order to measure performance relative to a baseline.

#machine learning #time series #open-source

A visual introduction to probability and statistics.

Statistics, is quickly becoming the most important and multi-disciplinary field of mathematics. According to the American Statistical Association, statistician is one of the top ten fastest-growing occupations and statistics is one of the fastest-growing bachelor degrees. Statistical literacy is essential to our data driven society. Yet, for all the increased importance and demand for statistical competence, the pedagogical approaches in statistics have barely changed. Using Mike Bostock’s data visualization software, D3.js, Seeing Theory visualizes the fundamental concepts covered in an introductory college statistics or Advanced Placement statistics class. Students are encouraged to use Seeing Theory as an additional resource to their textbook, professor and peers.

Also, it's open-source

#statistics #visualization

The world is awash in bullshit. […] We're sick of it. It's time to do something, and as educators, one constructive thing we know how to do is to teach people. So, the aim of this course is to help students navigate the bullshit-rich modern environment by identifying bullshit, seeing through it, and combatting it with effective analysis and argument.

Simplified, community-driven man pages!

Simpler man pages, oriented towards most common usage rather than exhaustiveness. E.g. for curl.

Exists as a website, a Dash docset, in command line…

Contribute to tldr here.

#cli #documentation

Numerical Optimization is one of the central techniques in Machine Learning. For many problems it is hard to figure out the best solution directly, but it is relatively easy to set up a loss function that measures how good a solution is - and then minimize the parameters of that function to find the solution.

#machine learning #math #visualization

Datalegreya is a typeface which can interweave data curves with text. E.g. b|1i|3n|2g|2o|1 |2h|2o|2p|3 becomes

Uses kerning, ligatures and contextual alternates

#font #typography #data #visualization

This project is targeting people who want to learn internals of ML algorithms or implement them from scratch. The code is much easier to follow than the optimized libraries and easier to play with. All algorithms are implemented in Python, using numpy, scipy and autograd.

#machine learning

As seen on TV! What if you could increase the resolution of your photos using technology from CSI laboratories? Thanks to deep learning and #NeuralEnhance, it's now possible to train a neural network to zoom in to your images at 2x or even 4x. You'll get even better results by increasing the number of neurons or training with a dataset similar to your low resolution image.

The catch? The neural network is hallucinating details based on its training from example images. It's not reconstructing your photo exactly as it would have been if it was HD. That's only possible in Hollywood — but using deep learning as "Creative AI" works and it is just as cool!

#machine learning #pictures

Imagine you have a flex container that wraps, containing 2 items with a flex-grow of respectfully 0 and 1 (and a . If the second one gets too large, it is moved to the next line and the first does not expand.

The fix:

.first { flex-grow: 1; } .second { flex-grow: 9999; flex-basis: 20em; }

#css

A deep learning software that easily generates images with a few brushstrokes (from UC Berkeley and Adobe CTL).

Available on GitHub

#machine learning

Trending Blogs

Recently Viewed Blogs

instanceof me