Sparse stochastic finite-state controllers for POMDPs
Bounded policy iteration is an approach to solving infinite-horizon POMDPs that represents policies as stochastic finite-state controllers and iteratively improves a controller by adjusting the parameters of each node using linear programming.
Our bounded policy iteration approach searches through the space of bounded-size, stochastic finite state controllers, combining sev- eral advantages of ...
Finite state controllers (FSCs) provide a simple, convenient way of representing policies for partially observable Markov decision processes (POMDPs).
Our bounded policy iteration approach searches through the space of bounded-size, stochastic finite state controllers, combining several advantages of gradient ...
We maintain a portfolio of research projects, providing individuals and teams the freedom to emphasize specific types of work. Our research philosophy.
Return to Article Details A Generic Technique for Synthesizing Bounded Finite-State Controllers Download Download PDF.
Feb 18, 2026 · Finite-state controllers (FSCs) are compact policy representations that map observations and internal memory to actions, ...
As a way to model DEC-POMDP policies with finite memory, finite state controllers provide an appealing solution. Each agent's policy can be represented as a.
In this paper, we propose a generic framework and related solver for the synthesis of bounded finite-state controllers. In particular, the solver is based ...
Bounded policy iteration (BPI) is an approach to solving infinite-horizon POMDPs that represents policies as sto- chastic finite-state controllers and ...