The problem with metrics is a big problem for AI

Re:The problem with metrics is a big problem for AI by sanxiyn on Saturday October 19, 2019 @01:18PM (#21299533)

The obvious answer is to learn the metrics itself too. OpenAI does interesting work in this area: "Deep reinforcement learning from human preferences", for example. https://arxiv.org/abs/1706.03741

[ Parent | on HN ]

Re:The problem with metrics is a big problem for AI by inimino on Saturday October 19, 2019 @05:20PM (#21300848)

I'm not sure if this is obvious, if what it means in practice is a metric that may be the wrong metric, but which is visible, being replaced by a metric that is invisible. At least in the first case we can reason about the metric and why optimizing for it might go wrong.

[ Parent | on HN ]

Re:The problem with metrics is a big problem for AI by AlexCoventry on Saturday October 19, 2019 @03:39PM (#21300310)

That's basically what GANs do, too. The recent results are amazing. https://learning-to-paint.github.io

[ Parent | on HN ]