Deep Successor Reinforcement Learning (2016)

Deep Successor Reinforcement Learning (2016)(arxiv.org)

69 points by aaronjg 9 years ago | 2 comments

skj 9 years ago |

It's not clear to me how this is interestingly different from model-based RL, where you learn the state function and reward function, and then use various types of simulation to learn a value function. I guess I'll have to read more than the abstract...

noelwelsh 9 years ago | |

Section 3.2 shows the successor representation (SR) definition. If I'm reading it correctly the SR might also be described as the discounted stationary distribution over states.

I haven't seen SR before in the RL literature, but the paper argues that this representation is useful for sub-goal identification. I guess I'll have to read more than the abstract as well :)