Reinforcement learning assumes an MDP with an a priori state space representation. Assume the state space is the raw images from a game, and we use convNets or another method to generate s latent state space. If we built the latent space first, we could run our RL algorithms on the fixed latent space representation, but what if we switched between learning a policy or value function on the latent state space, and also updating the latent space generator itself? Is there any theory or research on how to make such a system work?

  • $\begingroup$ See Embed to Control, a paper which learns a latent space and dynamics model which is locally linear, allowing for arbitrary control algorithms to operate in this space. $\endgroup$ – shimao Nov 30 '18 at 14:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.