- Template variable
X(U1,U2,...,Uk) is instantiated (duplicated) multiple times. - Template models:
- languages that specify how variables inherit dependency model from template
- Dynamic Bayesian networks
- Object-relational models:
- Directed: plate models
- Undirected
- Markov assumption
(X(t+1)⊥X(0:t−1)|X(t))
P(X(0:T))=P(X(0))∏t=0T−1P(X(t+1)|X(0:t))=P(X(0))∏t=0T−1P(X(t+1)|X(t)) - Time Invariance
For allt ,P(X(t+1)|X(t))=P(X′|X) - 2-time-slice Bayesian network
- A tranisition model (2TBN) over
X1,X2,…,Xn is specified as a BN fragment such that:
- The nodes include
X′1,X′2,…,X′n and a subset ofX1,X2,…,Xn - Only the nodes
X′1,X′2,…,X′n have parents and a CPD
- The nodes include
- The 2TBN defines a conditional distribution
P(X′|X)=∏i=1nP(X′i|PaX′i)
- A tranisition model (2TBN) over
- Dynamic Bayesian network
- A dynamic Bayesian network over
X1,…,Xn is defined by a
- 2TBN
BN→ overX1,…,Xn - a Bayesian network
BN(0) overX(0)1,…,X(0)n
- 2TBN
- A dynamic Bayesian network over
- Ground Network
For a trajectory over0,…,T we define a ground (unrolled network) such that
- The dependency model for
X(0)1,…,X(0)n is copied fromBN(0) - The dependency model for
X(t)1,…,X(t)n for allt>0 is copied fromBN→
- The dependency model for
Template model for Bayesian network
發表評論
所有評論
還沒有人評論,想成為第一個評論的人麼? 請在上方評論欄輸入並且點擊發布.