Assuring AI episode artwork

EPISODE · Dec 18, 2017 · 53 MIN

Assuring AI

from International Conference on Functional Programming 2017

John Launchbury, Chief Scientist of Galois Inc, gives the second keynote of the ICFP conference. Somewhat ironically, just as traditional systems software finally comes within reach of detailed mathematical verification, the demand for verification technology is rapidly expanding into domains that pose very different kinds of challenges. How could a face recognition technology be verified, for example, when it is not clear even how to state a correctness property? Or how should we attempt to verify an autonomous system like a self-driving car? This talk will attempt to lay out the problem space, and offer some ideas for how the verification community might start tackling the problem of assuring AI.

Episode metadata supplied by the publisher feed · Published Dec 18, 2017

Embed this episode

NOW PLAYING

Assuring AI

0:00 53:21

No transcript for this episode yet

We transcribe on demand. Request one and we'll notify you when it's ready — usually under 10 minutes.

No similar episodes found.

No similar podcasts found.

Frequently Asked Questions

How long is this episode of International Conference on Functional Programming 2017?

This episode is 53 minutes long.

When was this International Conference on Functional Programming 2017 episode published?

This episode was published on December 18, 2017.

Can I download this International Conference on Functional Programming 2017 episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.
URL copied to clipboard!