Factorial hidden Markov models

2022-08-28 — 2022-08-28

Wherein the hidden state is factorized into independent latent chains, and inference is rendered tractable by separable state variables, with an account of compatibility with neural architectures.

Bayes
dynamical systems
linear algebra
probability
signal processing
state space models
statistics
time series
Figure 1

Placeholder. Factorial HMMs factorize hidden state into independent (?) state variables. This obviously-nuts assumption makes them tractable but still expressive, which is what works for neural nets, so I guess we’re cool.

1 References

Ghahramani, and Jordan. 1996. “Factorial Hidden Markov Models.”
Schweiger, Erlich, and Carmi. 2019. “FactorialHMM: Fast and Exact Inference in Factorial Hidden Markov Models.” Bioinformatics.