Week 7 UAS

Question 1

What is Bayesian Network?

Answer 1

A data structure called a Bayesian network to represent the dependencies among variables. Bayesian networks
can represent essentially any full joint probability distribution and in many cases can do so very concisely.

Question 2

The difference between Naive Bayes and Bayesian Network

Answer 2

1. Naive Bayes, ignore correlation between variables
2. Bayesian Network, input variable can dependent

Bayesian Network also known as Belief Network, Probabilistic Network is graphic model for representation between variables.

Question 3

What is Markov Chain?

Answer 3

Markov chain is a simple type of stochastic process with many social science applications to solve optimization problem in stochastic modeling theory.

Question 4

Inference tasks?

Answer 4

1. Filtering, belief state, input to the decision process of a rational agent
2. Prediction, evaluation of possible action sequences; like filtering without the evidence
3. Smoothing, better estimate of past states, essential for learning
4. Most likely explanation, speech recognition, decoding with a noisy channel

Question 5

Forward-backward algorithm?

Answer 5

cache forward messages along the way time linear in t (polytree inference), space O(t|f|)

Question 6

most likely explanation using what algorithm?

Answer 6

viterbi algorithm

Question 7

Dynamic bayesian network?

Answer 7

1. Dynamic bayesian network are directed graphical models of stochastic process.
2. generalize HMM and KFL (kalman filter) by representing the hidden and observed state in terms of state variables.
3. Time-invariant
4. extension of BN to handle temporal models
5. compact parameterization to handle temporal models

Question 8

DBN vs HMM?

Answer 8

every HMM is a single-variable DBN; every discrete DBN is an HMM