AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Is God a Kitchen Sink of Approach to Reinforcement Learning?
I don't know to what extent it will be explicitly constructed. It might take unintuitive forms where you are telling the GPT like, hey, you have a declarative memory bank to which you can store and retrieve data from. That approach is something like humans. So God is very much a kitchen sink of approach to like reinforcement learning lots of different environments with a single fixed transformer model. I think it's a very sort of early result in that in that realm.