Duality Theory for Non-Markovian Linear Gaussian Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kudre, Aditya, Chang, Heng-Sheng, Mehta, Prashant G.
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910103334027264
author Kudre, Aditya
Chang, Heng-Sheng
Mehta, Prashant G.
author_facet Kudre, Aditya
Chang, Heng-Sheng
Mehta, Prashant G.
contents This work develops a duality theory for partially observed linear Gaussian models in discrete time. The state process evolves according to a causal but non-Markovian (or higher-order Gauss-Markov) structure, captured by a lower-triangular transition operator, which is related to transformer, with $T$ as the context length. The main contributions are: (i) a dual control system for the linear Gaussian model, formulated as a backward difference equation (B $Δ$ E); (ii) a duality principle establishing that a specific linear-quadratic optimal control problem for the B $Δ$ E is dual to the filtering problem for the partially observed model; and (iii) an explicit optimal control formula yielding a novel (transformer-like) linear predictor, referred to as the dual filter, whose computational complexity scales linearly in the time horizon $T$, in contrast to the $O(T^3)$ cost of classical smoothing and Wiener-Hopf approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2604_03909
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Duality Theory for Non-Markovian Linear Gaussian Models
Kudre, Aditya
Chang, Heng-Sheng
Mehta, Prashant G.
Systems and Control
Optimization and Control
This work develops a duality theory for partially observed linear Gaussian models in discrete time. The state process evolves according to a causal but non-Markovian (or higher-order Gauss-Markov) structure, captured by a lower-triangular transition operator, which is related to transformer, with $T$ as the context length. The main contributions are: (i) a dual control system for the linear Gaussian model, formulated as a backward difference equation (B $Δ$ E); (ii) a duality principle establishing that a specific linear-quadratic optimal control problem for the B $Δ$ E is dual to the filtering problem for the partially observed model; and (iii) an explicit optimal control formula yielding a novel (transformer-like) linear predictor, referred to as the dual filter, whose computational complexity scales linearly in the time horizon $T$, in contrast to the $O(T^3)$ cost of classical smoothing and Wiener-Hopf approaches.
title Duality Theory for Non-Markovian Linear Gaussian Models
topic Systems and Control
Optimization and Control
url https://arxiv.org/abs/2604.03909