The Gradient: Perspectives on AI cover image

Upol Ehsan on Human-Centered Explainable AI and Social Transparency

The Gradient: Perspectives on AI

00:00

The Four W's of Social Transparency

The four W's are going to get enough density depending on your news, right? Where they become large enough that you can feed back into the model. But the cool part is from day one, they're giving value to the user, right? So they are not like a grunt work of the dataset building task. They're actually building a corpus while getting value from it - which is very hard to achieve in any kind of dataset building tasks. The Y is actually giving insight into that crew knowledge. These are the variables that would be important for decision making, but sometimes are not captured in the AI's feature space.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app