Structured Probabilistic Models: Bayesian Networks and Beyond

8/6/98

Click here to start

Table of Contents

Structured Probabilistic Models: Bayesian Networks and Beyond

Probability distributions

Bayesian networks

BN inference (in theory)

BN inference (in practice)

Decision Making

Probabilistic models: applications

Uncertainty is unavoidable

Build on Strength

Model-based reasoning

Exploit domain structure

Problem: Knowledge engineering

Problem: Inference

Large-scale BNs

Exploit structure!

Beyond Bayesian networks

Inter-object interactions

Probabilistic model

Specific situation models

Inference: simple approach

Exploit structure

What have we gained…

Abstraction and refinement

Stochastic dynamic system

Dynamic Bayesian networks

Dynamic OOBNs: advantages

What about inference?

What about inference?

Unfortunately not…

Exploit structure

Revealed structure

Intelligent hospital manager

What do we need?

But we also need …

OOBNs ? probabilistic frames

Inter-frame dependencies

Structural uncertainty

Number Uncertainty

Reference uncertainty

Structural uncertainty: inference

Exploit structure

Exploit structure

Additional pieces

Author: koller

Email: koller@cs.stanford.edu

Home Page: http://robotics.stanford.edu/~koller