Planning as Dynamics Relaxation: Hippocampal Recurrent Network Realizes Optimal Goal-Directed Navigation
It is shown that a hippocampal network with appropriate recurrent connections can naturally achieve optimal goal-directed navigation via its relaxation dynamics, and it is proved that this field is mathematically equivalent to the desirability field of a Linearly-solvable Markov Decision Process (LMDP).
Yu-Hang He, Jun-Feng Zuo, Tian-Hao Chu et al.
· 0 citations