Deep Belief Markov Models for POMDP Inference

Giacomo Arcieri, Konstantinos G. Papakonstantinou, Daniel Straub +1

This work introduces a novel deep learning-based architecture, termed the Deep Belief Markov Model (DBMM), which provides efficient, model-formulation agnostic inference in Partial…