> When the state-of-the-art is 97% on a task, there's only so much room for impr... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		jean- on Jan 17, 2020 \| parent \| context \| favorite \| on: Researchers: Are we on the cusp of an ‘AI winter’? > When the state-of-the-art is 97% on a task, there's only so much room for improvement. When models commonly achieve 97% on a task, it means it's time to define a harder task, as it's long stopped providing any useful signal.

drongoking on Jan 18, 2020 [–]

I think when performance on a real world task can be expressed as a single percentage, we've over-simplified the hell out of it and it's time to rethink the problem.

Piskvorrr on Jan 18, 2020 | [–]

Or worse, overfitted - which means the solution will implode when faced with real data (RIP EH).

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact