Pub Date : 2006-09-01DOI: 10.1109/NSSPW.2006.4378817
L. Pavelková, M. Kárný, V. Šmídl
Linear state-space model with uniformly distributed innovations is considered. Its state and parameters are estimated under hard physical bounds. Off-line maximum a posteriori probability estimation reduces to linear programming. No approximation is required for sole estimation of either model parameters or states. The noise bounds are estimated in both cases. The algorithm is extended to: (i) on-line mode by estimating within a sliding window, and (ii) joint state and parameter estimation. This approach may be used as a starting point for full Bayesian treatment of distributions with restricted support.
{"title":"Towards Bayesian Filtering on Restricted Support","authors":"L. Pavelková, M. Kárný, V. Šmídl","doi":"10.1109/NSSPW.2006.4378817","DOIUrl":"https://doi.org/10.1109/NSSPW.2006.4378817","url":null,"abstract":"Linear state-space model with uniformly distributed innovations is considered. Its state and parameters are estimated under hard physical bounds. Off-line maximum a posteriori probability estimation reduces to linear programming. No approximation is required for sole estimation of either model parameters or states. The noise bounds are estimated in both cases. The algorithm is extended to: (i) on-line mode by estimating within a sliding window, and (ii) joint state and parameter estimation. This approach may be used as a starting point for full Bayesian treatment of distributions with restricted support.","PeriodicalId":388611,"journal":{"name":"2006 IEEE Nonlinear Statistical Signal Processing Workshop","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116205628","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-09-01DOI: 10.1109/NSSPW.2006.4378820
M. Briers, A. Doucet, S. Singh, K. Weekes
This paper discloses a novel algorithm for efficient inference in undirected graphical models using Sequential Monte Carlo (SMC) based numerical approximation techniques. The developed methodology extends the applicability of the much celebrated Loopy Belief Propagation (LBP) algorithm to nonlinear, non-Gaussian models, whilst retaining a computational cost that is linear in the number of sample points (or particles). The work presented is thus a general framework that can be applied to a plethora of novel non-linear signal processing problems. In this paper, we apply our inference algorithm to the (sequential problem of) articulated object tracking.
{"title":"Particle Filters for Graphical Models","authors":"M. Briers, A. Doucet, S. Singh, K. Weekes","doi":"10.1109/NSSPW.2006.4378820","DOIUrl":"https://doi.org/10.1109/NSSPW.2006.4378820","url":null,"abstract":"This paper discloses a novel algorithm for efficient inference in undirected graphical models using Sequential Monte Carlo (SMC) based numerical approximation techniques. The developed methodology extends the applicability of the much celebrated Loopy Belief Propagation (LBP) algorithm to nonlinear, non-Gaussian models, whilst retaining a computational cost that is linear in the number of sample points (or particles). The work presented is thus a general framework that can be applied to a plethora of novel non-linear signal processing problems. In this paper, we apply our inference algorithm to the (sequential problem of) articulated object tracking.","PeriodicalId":388611,"journal":{"name":"2006 IEEE Nonlinear Statistical Signal Processing Workshop","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122228791","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-09-01DOI: 10.1109/NSSPW.2006.4378855
A. Cemgil
Conditional Gaussian changepoint models are an interesting subclass of jump-Markov dynamic linear systems, in which, unlike the majority of such intractable hybrid models, exact inference is achievable in polynomial time. However, many applications of interest involve several simultaneously unfolding processes with occasional regime switches and shared observations. In such scenarios, a factorial model, where each process is modelled by a changepoint model is more natural. In this paper, we derive a sequential Monte Carlo algorithm, reminiscent to the Mixture Kalman filter (MKF) [1]. However, unlike MKF, the factorial structure of our model prohibits the computation of the posterior filtering density (the optimal proposal distribution). Even evaluating the likelihood conditioned on a few switch configurations can be time consuming. Therefore, we derive a propagation algorithm (upward-downward) that exploits the factorial structure of the model and facilitates computing Kalman filtering recursions in information form without the need for inverting large matrices. To motivate the utility of the model, we illustrate our approach on a large model for polyphonic pitch tracking.
{"title":"Sequential Inference for Factorial Changepoint Models","authors":"A. Cemgil","doi":"10.1109/NSSPW.2006.4378855","DOIUrl":"https://doi.org/10.1109/NSSPW.2006.4378855","url":null,"abstract":"Conditional Gaussian changepoint models are an interesting subclass of jump-Markov dynamic linear systems, in which, unlike the majority of such intractable hybrid models, exact inference is achievable in polynomial time. However, many applications of interest involve several simultaneously unfolding processes with occasional regime switches and shared observations. In such scenarios, a factorial model, where each process is modelled by a changepoint model is more natural. In this paper, we derive a sequential Monte Carlo algorithm, reminiscent to the Mixture Kalman filter (MKF) [1]. However, unlike MKF, the factorial structure of our model prohibits the computation of the posterior filtering density (the optimal proposal distribution). Even evaluating the likelihood conditioned on a few switch configurations can be time consuming. Therefore, we derive a propagation algorithm (upward-downward) that exploits the factorial structure of the model and facilitates computing Kalman filtering recursions in information form without the need for inverting large matrices. To motivate the utility of the model, we illustrate our approach on a large model for polyphonic pitch tracking.","PeriodicalId":388611,"journal":{"name":"2006 IEEE Nonlinear Statistical Signal Processing Workshop","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130358572","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-09-01DOI: 10.1109/NSSPW.2006.4378835
G. Ueno, T. Higuchi, T. Kagimoto, N. Hirose
We report the first application of the ensemble Kalman filter (EnKF) to an intermediate coupled atmosphere-ocean model by [1], into which the sea surface height (SSH) anomaly observations by TOPEX/POSEIDON (T/P) altimetry are assimilated. Smoothed estimates ofthe 54,403 dimensional state are obtained from 1981 observational points with 2048 ensemble members. While data assimilated are SSH anomalies alone, an ensemble experiment of 2002/03 El Niño event based on the EnKF can predict consistent Niño 3 sea surface temperature (SST) anomalies about 5 months in advance.
{"title":"Application of the Ensemble Kalman Filter to Atmosphere-Ocean Coupled Model","authors":"G. Ueno, T. Higuchi, T. Kagimoto, N. Hirose","doi":"10.1109/NSSPW.2006.4378835","DOIUrl":"https://doi.org/10.1109/NSSPW.2006.4378835","url":null,"abstract":"We report the first application of the ensemble Kalman filter (EnKF) to an intermediate coupled atmosphere-ocean model by [1], into which the sea surface height (SSH) anomaly observations by TOPEX/POSEIDON (T/P) altimetry are assimilated. Smoothed estimates ofthe 54,403 dimensional state are obtained from 1981 observational points with 2048 ensemble members. While data assimilated are SSH anomalies alone, an ensemble experiment of 2002/03 El Niño event based on the EnKF can predict consistent Niño 3 sea surface temperature (SST) anomalies about 5 months in advance.","PeriodicalId":388611,"journal":{"name":"2006 IEEE Nonlinear Statistical Signal Processing Workshop","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121510416","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-09-01DOI: 10.1109/NSSPW.2006.4378850
P. Djurić, M. Bugallo
Cost-reference particle filtering (CRPF) is a methodology for recursive estimation of unobserved states of dynamic systems using a stream of particles and their associated costs. It is similar to the standard particle filtering (SPF) methodology in that it is comprised of similar steps, that is, (1) propagation of particles, (2) cost (weight) computation, and (3) resampling. The main difference between CRPF and SPF is that the former uses very mild statistical assumptions and the latter is based on strong probabilistic assumptions. In problems where some of the states are linear given the rest of the states, one can employ an SPF scheme with improved filtering performance. In the literature on SPF, this methodology is known as Rao-Blackwellized particle filtering. In this paper, we show how we can exploit a similar idea in the context of CRPF.
{"title":"Cost-Reference Particle Filtering for Dynamic Systems with Nonlinear and Conditionally Linear States","authors":"P. Djurić, M. Bugallo","doi":"10.1109/NSSPW.2006.4378850","DOIUrl":"https://doi.org/10.1109/NSSPW.2006.4378850","url":null,"abstract":"Cost-reference particle filtering (CRPF) is a methodology for recursive estimation of unobserved states of dynamic systems using a stream of particles and their associated costs. It is similar to the standard particle filtering (SPF) methodology in that it is comprised of similar steps, that is, (1) propagation of particles, (2) cost (weight) computation, and (3) resampling. The main difference between CRPF and SPF is that the former uses very mild statistical assumptions and the latter is based on strong probabilistic assumptions. In problems where some of the states are linear given the rest of the states, one can employ an SPF scheme with improved filtering performance. In the literature on SPF, this methodology is known as Rao-Blackwellized particle filtering. In this paper, we show how we can exploit a similar idea in the context of CRPF.","PeriodicalId":388611,"journal":{"name":"2006 IEEE Nonlinear Statistical Signal Processing Workshop","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121718219","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-11-16DOI: 10.1109/NSSPW.2006.4378840
G. Jóhannesson, K. Dyer, W. Hanley, B. Kosović, S. Larsen, G. Loosmore, J. Lundquist, A. Mirin
The release of hazardous materials into the atmosphere can have a tremendous impact on dense populations. We propose an atmospheric event reconstruction framework that couples observed data and predictive computer-intensive dispersion models via Bayesian methodology. Due to the complexity of the model framework, a sampling-based approach is taken for posterior inference that combines Markov chain Monte Carlo (MCMC) and sequential Monte Carlo (SMC) strategies.
{"title":"Sequential Monte-Carlo Framework for Dynamic Data-Driven Event Reconstruction for Atmospheric Release","authors":"G. Jóhannesson, K. Dyer, W. Hanley, B. Kosović, S. Larsen, G. Loosmore, J. Lundquist, A. Mirin","doi":"10.1109/NSSPW.2006.4378840","DOIUrl":"https://doi.org/10.1109/NSSPW.2006.4378840","url":null,"abstract":"The release of hazardous materials into the atmosphere can have a tremendous impact on dense populations. We propose an atmospheric event reconstruction framework that couples observed data and predictive computer-intensive dispersion models via Bayesian methodology. Due to the complexity of the model framework, a sampling-based approach is taken for posterior inference that combines Markov chain Monte Carlo (MCMC) and sequential Monte Carlo (SMC) strategies.","PeriodicalId":388611,"journal":{"name":"2006 IEEE Nonlinear Statistical Signal Processing Workshop","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132840959","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：