Stochastic Processes And Filtering Theory

By Florian Nadel. 2016

1,472 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Deep Kalman Filters (2015, methodology): “For example, the Extended Kalman Filter (Jazwinski, 2007) and the Unscented Kalman Filter (Wan et al. , 2000) are two different methods to learn temporal models with non-linear transition and emission distributions (see also Roweis & Ghahramani (2000) and Haykin (2004)).”
Data assimilation: making sense of Earth Observation (2014, methodology): “The KF smoother version is called the Kalman smoother (Jazwinski, 1970).”
AutoRally: An Open Platform for Aggressive Autonomous Driving (2019, unknown): “…estimator for S i is unavailable and many suboptimal schemes are either too restrictive for nonlinear applications or too computationally demanding [30], [31], this work adopts the adaptive limited memory algorithm in [29], with the following two extensions: 1) The algorithm is developed for a…”
Using Inertial Sensors for Position and Orientation Estimation (2017, unknown): “The latter approach is closely related to iterated Kalman smoothers [10, 65].”
TRADI: Tracking deep neural network weight distributions (2019, background): “Bayesian recursive filtering [3] aims to estimate the state of a hidden Markov process, observed through a state space equations system.”
A review of operational methods of variational and ensemble‐variational data assimilation (2017, unknown): “Substituting Pb(N) into the Kalman filter (KF) update equation (Jazwinski, 1970; Evensen, 2003) gives: xa(i) = xb(i)+PbHT ( R + HPbHT )−1 d(i), ≈ xb(i)+Xb(HXb)T { R+HXb(HXb)T }−1 d(i), (28) where d(i) = yo(i) − H(xb(i)).”
On the representation error in data assimilation (2018, unknown): “If we correct only the bias, the analysis-error statistics calculated through either the Schmidt–Kalman filter or the traditional filter would not be correct and therefore, in addition, evolution equations for the correlation between the error and the state would need to be included (Jazwinski, 1970”
Learning Stationary Time Series using Gaussian Processes with Nonparametric Kernels (2015, methodology): “…of this paper is to devise a computationally tractable method for learning the filter h(t) (known as system identification in the control community [17]) and inferring the white-noise process x(t) from a noisy dataset y ∈ R N produced by their convolution and additive Gaussian noise, y(t) = f (t)…”