Lesson 11 / 12: Markov Chain | Hidden Markov Chain

Planning

determining a sequence of actions that are known to achieve a particular objective when performed

Markov Chains

defined by the transition probabilities of the events happening, given the possible previous states

Markov Chains

special class of graphical models where every node has one child node and there is a chain of these nodes

Markov Chains

basic conditional independence

Growable Bayesian Network

chain is just

Sequence of states

Markov chain’s reasoning is based solely from the _ _ _

Hidden Markov Models

a sequence of random variables Z1, Zt, such that the distribution of Zt depends only on the hidden state xt of an associated Markov Chain

P(x1)

initial distribution

P(x1 | Xt-1)

transition probabilities

P(Zt | xt)

observation probabilities

Markov Hidden Process

future depends on past via the present

Independent

current observation is _ of all other observations given the current state