10-803 Markov Logic Networks

1
10-803Markov Logic Networks

• Instructor
• Pedro Domingos

2
Logistics

• Instructor Pedro Domingos
• Email pedrod_at_cs.washington.edu
• Office Wean 5317
• Office hours Thursdays 200-300
• Course secretary Sharon Cavlovich
• Web http//www.cs.washington.edu/homes/pedrod/80
  3/
• Mailing list 10803-students_at_cs.cmu.edu

3
Source Materials

• TextbookP. Domingos D. Lowd,Markov Logic An
  Interface Layer for AI, Morgan Claypool, 2008
• Papers
• SoftwareAlchemy (alchemy.cs.washington.edu)
• MLNs, datasets, etc.Alchemy Web site

4
Project

• Possible projects
• Apply MLNs to problem youre interested in
• Develop new MLN algorithms
• Other
• Key dates/deliverables
• This week Download Alchemy and start playing
• October 9 (preferably earlier) Project proposal
• November 6 Progress report
• December 4 Final report and short presentation
• Winter 2009 Conference submission (!)

5
What Is Markov Logic?

• A unified language for AI/ML
• Special cases
• First-order logic
• Probabilistic models
• Syntax Weighted first-order formulas
• Semantics Templates for Markov nets
• Inference Logical and probabilistic
• Learning Statistical and ILP

6
Why Take this Class?

• Powerful set of conceptual tools
• New way to look at AI/ML
• Powerful set of software tools
• Increase your productivity
• Attempt more ambitious applications
• Powerful platform for developing new learning and
  inference algorithms
• Many fascinating research problems

Caveat Not mature!
7
Sample Applications

• Information extraction
• Entity resolution
• Link prediction
• Collective classification
• Web mining
• Natural language processing

• Computational biology
• Social network analysis
• Robot mapping
• Activity recognition
• Personal assistants
• Probabilistic KBs
• Etc.

8
Overview of the Class

• Background
• Markov logic
• Inference
• Learning
• Extensions
• Your projects

9
Background

• Markov networks
• Representation
• Inference
• Learning
• First-order logic
• Representation
• Inference
• Learning (a.k.a. inductive logic programming)

10
Markov Logic

• Representation
• Properties
• Relation to first-order logic and statistical
  models
• Related approaches

11
Inference

• Basic MAP and conditional inference
• The MC-SAT algorithm
• Knowledge-based model construction
• Lazy inference
• Lifted inference

12
Learning

• Weight learning
• Generative
• Discriminative
• Incomplete data
• Structure learning and theory revision
• Statistical predicate invention
• Transfer learning

13
Extensions

• Continuous domains
• Infinite domains
• Recursive MLNs
• Relational decision theory

14
Your Projects

• (TBA)

15
Class begins here.
16
AI The First 100 Years
IQ
Human Intelligence
Artificial Intelligence
1956
2056
2006
17
AI The First 100 Years
IQ
Human Intelligence
Artificial Intelligence
1956
2056
2006
18
AI The First 100 Years
Artificial Intelligence
IQ
Human Intelligence
1956
2056
2006
19
The Interface Layer
Applications
Interface Layer
Infrastructure
20
Networking
WWW
Email
Applications
Internet
Interface Layer
Protocols
Infrastructure
Routers
21
Databases
ERP
CRM
Applications
OLTP
Interface Layer
Relational Model
Transaction Management
Infrastructure
Query Optimization
22
Programming Systems
Programming
Applications
Interface Layer
High-Level Languages
Compilers
Code Optimizers
Infrastructure
23
Hardware
Computer-Aided Chip Design
Applications
Interface Layer
VLSI Design
Infrastructure
VLSI modules
24
Architecture
Operating Systems
Applications
Compilers
Interface Layer
Microprocessors
ALUs
Infrastructure
Buses
25
Operating Systems
Applications
Software
Interface Layer
Virtual machines
Infrastructure
Hardware
26
Human-Computer Interaction
Applications
Productivity Suites
Interface Layer
Graphical User Interfaces
Infrastructure
Widget Toolkits
27
Artificial Intelligence
Planning
Robotics
Applications
NLP
Multi-Agent Systems
Vision
Interface Layer
Representation
Inference
Infrastructure
Learning
28
Artificial Intelligence
Planning
Robotics
Applications
NLP
Multi-Agent Systems
Vision
Interface Layer
First-Order Logic?
Representation
Inference
Infrastructure
Learning
29
Artificial Intelligence
Planning
Robotics
Applications
NLP
Multi-Agent Systems
Vision
Interface Layer
Graphical Models?
Representation
Inference
Infrastructure
Learning
30
Logical and Statistical AI
31
We Need to Unify the Two

• The real world is complex and uncertain
• Logic handles complexity
• Probability handles uncertainty

32
Artificial Intelligence
Planning
Robotics
Applications
NLP
Multi-Agent Systems
Vision
Interface Layer
Markov Logic
Representation
Inference
Infrastructure
Learning