Welcome to Quantum Magazine's podcast.

Each episode we bring you stories about developments in science and mathematics.

I'm Susan Vallett.

Imagine that your neighbor calls to ask a favor.

Could you please feed their pet rabbit some carrot slices?

Easy enough, you'd think.

You can imagine their kitchen, even if you've never been there.

Carrots in a fridge, a drawer holding various knives.

It's abstract knowledge.

You don't know what your neighbor's carrots and knives look like exactly,

but you won't take a spoon to a cucumber.

Artificial intelligence programs can't compete.

But what happens if we pre-trained them to understand these concepts?

That's next.

Quantum Magazine is an editorially independent online publication supported by the

Simon's Foundation to enhance public understanding of science.

What seems to you like an easy task is a huge undertaking for current algorithms.

An AI-trained robot can find a specified knife and carrot hiding in a familiar kitchen.

But in a different kitchen, it will lack the abstract skills to succeed.

Victor Jone is a graduate student in computer science at the University of Washington.

When you have these reinforced learning systems, they are very difficult to train and

they don't generalize to new environments.

...