Algorithmic art shows what the machine sees

Tom White is an artist who uses neural networks to draw abstract pictures of objects. What looks blobby and fuzzy to us looks more concrete to the machine.

James Vincent for The Verge:

That “voice” is actually a series of algorithms that White has dubbed his “Perception Engines.” They take the data that machine vision algorithms are trained on — databases of thousands of pictures of objects — and distill it into abstract shapes. These shapes are then fed back into the same algorithms to see if they’re recognized. If not, the image is tweaked and sent back, again and again, until it is. It’s a trial and error process that essentially ends up reverse-engineering the algorithm’s understanding of the world.

Tags: ,

Cartography Playground

Map-making is a tricky business with many variables to consider that can directly change how someone interprets the land and people in a location. The Cartography Playground is a simple site to test these variables interactively. Learn about algorithms, mess with appearance, and toggle through representations.

Tags: ,

Controlling the Unaccountable Algorithm

BBC Radio 4 looks at algorithms in our everyday lives and why we should care what goes on in the black box.

Algorithms are the powerful mathematical tools which shape so much of modern life, from the news which appears in our timelines to the adverts which pop up on our computer. But with algorithms now assessing CVs for jobs, or mortgage applications, the need to understand what they do, and if necessary challenge them is greater than ever before. So how do we exert legal or democratic control over the Unaccountable Algorithm? Emily Bell investigates.

Tags: ,

Algorithms for the Traveling Salesman Problem visualized

The Traveling Salesman Problem is a popular puzzle that asks for the shortest route between a set of points such that you visit each point once and end up back where you started. The problem is trivial for a few points, but it gets tricky as you add more. Here are are a few of the strategies in action.

See also this interactive simulation.

Or, you can try using genetic algorithms. [via kottke]

Tags: , ,

Placement matching algorithms

Students want to get into a school, and schools want certain students. Match. Med students want to get into a specific residency program, and certain programs want specific students. Match.

Tim Harford explains the role of matching algorithms to make picking fair for all parties. The process gets messy when you start looking at thousands of individuals and organizations with multiple preferences each.

The deferred acceptance algorithm is just the start of a successful market design, because details matter. In New York City, there are different application procedures for certain specialised schools. When assigning hospital residencies, the US National Resident Matching Program needed to cope with pairs of romantically attached doctors who wanted two job offers in the same city. These complexities sometimes mean there is no perfect matching algorithm, and the challenge is to find a system that is good enough to work.

Humans.

Tags: , ,

Inadvertent algorithmic cruelty

If you logged into Facebook the past couple of weeks, you saw your friends' automatically generated year-end reviews. Estimated events and popular pictures appear in chronological order. Facebook eventually pinned your own year in review at the top of your feed for perusal. Seems harmless — until you realize there are people who don't want to look back, like Eric Meyer, whose daughter died this year.

And I know, of course, that this is not a deliberate assault. This inadvertent algorithmic cruelty is the result of code that works in the overwhelming majority of cases, reminding people of the awesomeness of their years, showing them selfies at a party or whale spouts from sailing boats or the marina outside their vacation house.

But for those of us who lived through the death of loved ones, or spent extended time in the hospital, or were hit by divorce or losing a job or any one of a hundred crises, we might not want another look at this past year.

See also Meyer's follow-up. While many took the original post as a way to hate on Facebook, Meyer didn't mean it like that.

Tags: , ,

Visualizing algorithms

Starry Night sampled

Mike Bostock, who you might recognize from such things as Data-Driven Documents or the New York Times, writes on the value of visualizing algorithms for entertaining, teaching, learning, and debugging.

Algorithms are a fascinating use case for visualization. To visualize an algorithm, we don’t merely fit data to a chart; there is no primary dataset. Instead there are logical rules that describe behavior. This may be why algorithm visualizations are so unusual, as designers experiment with novel forms to better communicate. This is reason enough to study them.

But algorithms are also a reminder that visualization is more than a tool for finding patterns in data. Visualization leverages the human visual system to augment human intellect: we can use it to better understand these important abstract processes, and perhaps other things, too.

At the very least, you'll have fun scrolling through the animated visuals that show how various algorithms work, but read the whole thing. It's good.