[ExI] DeepMind AI teaches itself about the world by watching videos

BillK pharos at gmail.com
Fri Aug 11 15:34:12 UTC 2017

Daily news    10 August 2017
By Matt Reynolds


“We want to build machines that continuously learn about their
environment in an autonomous manner,” says Pulkit Agrawal at the
University of California, Berkeley. Agrawal, who wasn’t involved with
the work, says this project takes us closer to the goal of creating AI
that can teach itself by watching and listening to the world around

While the AI in the DeepMind project doesn’t interact with the real
world, Agarwal says that perfecting self-supervised learning will
eventually let us create AI that can operate in the real world and
learn from what it sees and hears.

But until we reach that point, self-supervised learning might be a
good way of training image and audio recognition algorithms without
input from vast amounts of human-labelled data. The DeepMind algorithm
can correctly categorise an audio clip nearly 80 per cent of the time,
making it better at audio-recognition than many algorithms trained on
labelled data.

Such promising results suggest that similar algorithms might be able
to learn something by crunching through huge unlabelled datasets like
YouTube’s millions of online videos. “Most of the data in the world is
unlabelled and therefore it makes sense to develop systems that can
learn from unlabelled data,” Agrawal says.


Maybe it's just me, but letting AI learn how best to survive in the
world by watching videos on the web sounds like a disaster waiting to

What is the AI going to select as the 'best' role models?  Trump??
Stalin????  Pop Eye????


More information about the extropy-chat mailing list