State Space Models: A Modern Approach

State Space Models: A Modern Approach

Scratchpad
What are State Space Models?
Inference in discrete SSMs
Inference in linear-Gaussian SSMs
Extended (linearized) methods
Unscented methods
- Unscented filtering
- Unscented smoothing
Quadrature and cubature methods
Posterior linearization
Assumed Density Filtering
Variational inference
Particle filtering
Sequential Monte Carlo
Offline parameter estimation (learning)
Multi-target tracking
Data assimilation using Ensemble Kalman filter
Bayesian non-parametric SSMs
Changepoint detection
Timeseries forecasting
Markovian Gaussian processes
Differential equations and SSMs
Optimal control
Bibliography

Powered by Jupyter Book

Inference in discrete SSMs

Inference in discrete SSMs¶

This chapter covers inference in HMMs.

Hidden Markov Models
HMM filtering (forwards algorithm)
HMM smoothing (forwards-backwards algorithm)
Viterbi algorithm
Parallel HMM smoothing
Forwards-filtering backwards-sampling algorithm

See (sec:casino-ex).

previous

What are State Space Models?

next

Hidden Markov Models

By Kevin Murphy, Scott Linderman, et al.
© Copyright 2021.