Fetching the paper…

Learning Good State and Action Representations via Tensor Decomposition · Around