AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Drawbacks of Reinforcement Learning With Human Feedback
There are things missing in the model, and that's what I've been focused on. One reason is because humans separate knowledge of the world from how to take decisions. In a goal playing system like AlphaGo, for example, you don't need to interact with the rest of the world. The machinery to take that information and turn it into good decisions is very complex. That's why we need really, really large neural nets to do the inference.