'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!

Or

Sign up to gain exclusive access to email subscriptions, event invitations, competitions, giveaways, and much more.

Membership is free, and your security and privacy remain protected. View our privacy policy before signing up.

Error: Please check your email address.
Show Comments

Blog Posts

Searching for social and marketing data

Many marketers, agencies - and everyone in between - get caught up on bubble references and data points. They’ll use Facebook best practice as the only best practice for Facebook executions and only consider metrics and responses of the one channel they’re expected to deliver on.

Isaac Lai

Connections strategy lead, VMLY&R Sydney

Why Australia needs more leaders

A few weeks ago, our Prime Minister, Scott Morrison took it upon himself to tell companies and their CEOs where to go when it came to societal issues. It wasn’t an organisation’s place to get involved. Instead, he said it should be left to governments to solve societies challenges.

Dan Banyard

Managing director, Edentify

3 skills you need to drive better collaboration

A study published in The Harvard Business Review found the time spent in collaborative activities at work has increased by over 50 per cent in the past two decades. Larger projects; complicated problems; tighter timeframes: These require bigger teams with specialised skillsets and diverse backgrounds, often dispersed globally.

Jen Jackson

CEO, Everyday Massive

Informative blog. Xero is a well-known revolutionized accounting software, specifically developed to provide best User Experience and mak...

NavkarConsultancyServices

Xero evolves to fit a changing marketplace

Read more

>Writes article about how to show diversity in an authentic way>All featured opinions are from white women

Jennifer Metcalfe

Food for Thought: How can brands show diversity in an authentic way?

Read more

Excellent post, congratulations !!! - Prof Paulo Coelho | https://www.drpaulocoelho.c...

Prof Paulo Coelho

The B2C and B2B marketing transformation helping Invisalign win more smiles

Read more

Great article - but regarding "For a team to achieve their full potential, Edmonson also advises leaders balance psychological safety wi...

Sim

3 skills you need to drive better collaboration - Business leadership - CMO Australia

Read more

I would invest money in machine learning. I think it's important to do that. Don't you agree?

Polly Valentine

Metcash to use AI for promotional planning optimisation in liquor division

Read more

Latest Podcast

More podcasts

Sign in