[defen]

> In order to decrease the number of samples needed from the environment, we expect that ideas related to model-based RL — where you spend a huge number of neural network flops to learn a model of the environment — will be the way to go.

Will those models be introspectible / transferrable? One thing I'm curious about is how AI's learn about novel actions / scenarios which are "fatal" in the real world? Humans generally spend a lot of time being taught these things (rather than finding out for themselves obviously) and eventually come up with a fairly good set of rules about how not to die in stupid ways.

[wassyape]

Transferability depends on the way the models is set up, and moves on a scale.

Introspectable: given that you can ask unlimited "What if" questions models, we should be able to get a lot of insights into how the models work internally. And you can often design them to be introspectable as some performance or complexity cost. (if that's what you meant by introspectable).