'Ex Machina', here we come: A new algorithm helps computers learn the way we do

In a 'visual Turing test,' human judges couldn't tell them apart

Machine learning is all about getting computers to "understand" new concepts, but it's still a pretty inefficient process, often requiring hundreds of examples for training. That may soon change, however, thanks to new research published on Friday.

Aiming to shorten the learning process and make it more like the way humans acquire and apply new knowledge based on just a few examples, a team of researchers has developed what they call a Bayesian Program Learning framework and then used it to teach computers to identify and reproduce handwritten characters based on just a single example.

Whereas standard pattern-recognition algorithms represent concepts as configurations of pixels or collections of features, the BPL approach learns by “explaining” the data provided to the algorithm -- in this case, the sample character. Concepts are represented as probabilistic computer programs and the algorithm essentially programs itself by constructing code to produce the letter it sees. It can also capture variations in the way different people draw a given letter.

The model also “learns to learn” by using knowledge from previous concepts to speed learning on new ones, so it can use knowledge of the Latin alphabet to learn letters in the Greek alphabet more quickly, for example.

Most compelling of all is that the algorithm allowed computers to pass a sort of "visual Turing test." Specifically, the researchers asked both humans and computers to reproduce a series of handwritten characters after being shown just a single example of each; in some cases, subjects were asked to create entirely new characters in the style of those originally shown. Bottom line: human judges couldn't tell the results apart.

The researchers have applied their model to more than 1,600 types of handwritten characters in 50 writing systems, including Sanskrit, Tibetan, Gujarati and Glagolitic. They even tried it on invented characters such as those from the television series "Futurama."

A paper describing the research was published Friday in the journal Science. Its authors were Brenden Lake, a Moore-Sloan Data Science Fellow at New York University; Ruslan Salakhutdinov, an assistant professor of Computer Science at the University of Toronto; and Joshua Tenenbaum, a professor at MIT in the Department of Brain and Cognitive Sciences and the Center for Brains, Minds and Machines.

“It has been very difficult to build machines that require as little data as humans when learning a new concept,” said Salakhutdinov. “Replicating these abilities is an exciting area of research connecting machine learning, statistics, computer vision, and cognitive science.”

Join the newsletter!

Error: Please check your email address.
Show Comments

Blog Posts

To DMP or not to DMP?

There are plenty of brands that can benefit from plugging into a data management platform. But should you engage an agency to run one or bring it in-house?

Ben Willee and Richard Taylor

Spinach Advertising

Innovations in retail will bring creative and technology closer than ever

While approaching a customer in a shop and asking what you can help them with is Retail 101, how many of us actually enjoy being approached? Generally, you have to give the forced, fake smile and say, “Just browsing, thanks,” while screaming on the inside, “just leave me alone!” Maybe it’s just me?

Jason Dooris

CEO and founder, Atomic 212

There’s a brand in my digital soup

Not a day passes by in the life of business executives where digital innovation or the prospect of disruption is not front of mind. This in turn, drives an unrelenting flow of questioning, discussion and strategy papers.

Jean-Luc Ambrosi

Author, marketer

It is really inspiring to see that medial psychology & machines are going hand in hand to innovate new things and even are improving ...

CBT Professionals

How psychology is shaping better machine learning

Read more

Anything is achievable when you talk the talk .. good luck

Mo Al Hooti

Data Creative crowns Matt Bates as new CMO

Read more

Great hire and good luck Andrew. I am sure you'll do a fabulous jobRob

Rob

Amaysim marketing and commercial chief joins property tech investment startup

Read more

I need to be reborn before i can grow up & become a contestent.

Kaye Peterkin

Channel Nine's content now streamed digitally on 9Now

Read more

I hope The Block, will continue for many many more seasons.When i die, & grow up, I want yo come back as a contestent.

Kaye Peterkin

Channel Nine's content now streamed digitally on 9Now

Read more

Latest Podcast

More podcasts

Sign in