AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Is There a Relationship Between Explainability Research and Chess?
Ano: I think there's a lot of potential to adopt some of the techniques that people have developed to explain what's going on inside models. And if i could explain better by poking into a models and understanding what they understand innately, than maybe i can advise or guide a teacher or coach in a better way. We did some interesting work at some point where we're trying to reflect the actual problistic uncertainties that the machine had internally in various models,. In facial expressions. This isn't the case of a virtual avatar that sits on a screen and tries to interact with people. This is something we often do consciously and unitself. Ano: So it's interesting,