'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
cmo-xs-promo

Latest Videos

More Videos

Undoubtedly, Robotic Process Automation enables marketing teams to accomplish repetitive tasks with fewer resources at a great accuracy.

Raj

What marketers need to know about robotic process automation

Read more

who wants to date me I am 9 years old and i am a boy

Jeremy Hawkins

Sink a sub gaming experience signals Subway's renewed brand push

Read more

Great read. I agree that it should be a perfect balance between interacting with your customers and knowing your brand. As a business, yo...

Caroline Scott

7 ways CMOs can improve their customer engagement game

Read more

Very true. Team development helps improve collaboration among the team members. I was able to improve my team's collaboration skills by t...

Quent Sinder

Why empowering others can help make you a great leader

Read more

CRM is a very good software that can help you succeed in your business. In my company, this system has allowed me to improve customer rel...

Anna Janicka

Sensis rebrands to Thryv and brings business software to Australian SMBs

Read more

Blog Posts

Unboxing 101 - How savvy influencer engagement can build a brand

The humble unboxing video is a powerful tool. Correctly executed, it harnesses consumer fandom, viral authenticity and brand design magic to deliver a high-impact message to a tightly targeted cohort of consumers.

Gali Arnon

Chief marketing officer, Fiverr

​Power to the people

Purpose is the ultimate statement of intent for many organisations. Why are we here? What are we trying to achieve?

Rich Curtis

CEO, FutureBrand A/NZ

The playbook to develop strategic brand moats

Warren Buffet is an unlikely ally for marketers. But his belief businesses need strategic moats that increase their value in the market while acting as barriers to competitors can offer marketers a new playbook for brand building and driving growth.

Fabian Di Marco

Founder and managing director, Tzu & Co

Sign in