Kenneth Czuprynski

2 accepted papers

2022

Scalable Gradient Ascent for Controllers in Constrained POMDPs

ICRA 2022poster

This paper presents a novel gradient ascent al-gorithm and nonlinear programming algorithm for finite state controller policies in constrained partially observable Markov decision processes (CPOMDPs). A key component of the gradient ascent algorithm is a constraint projection to ensure constraints a…

Cited by 9 Source ↗Scholar ↗

2021

Scalable POMDP Decision-Making Using Circulant Controllers

Kyle Hollins Wray, Kenneth Czuprynski

ICRA 2021poster

This paper presents a novel policy representation for partially observable Markov decision processes (POMDPs) called circulant controllers and a provably efficient gradient-based algorithm for them. A formal mathematical description is provided that leverages circulant matrices for the controller’s…

Cited by 2 Source ↗Scholar ↗