New Complexity Results for MAP in Bayesian Networks

Similar documents
Bayesian Networks. Mausam (Slides by UW-AI faculty)

Credal Networks for Operational Risk Measurement and Management

Bayesian Networks. Read R&N Ch Next lecture: Read R&N

Approximating the Partition Function by Deleting and then Correcting for Model Edges

Big Data, Machine Learning, Causal Models

Course: Model, Learning, and Inference: Lecture 5

Large-Sample Learning of Bayesian Networks is NP-Hard

Bayesian networks - Time-series models - Apache Spark & Scala

Finding the M Most Probable Configurations Using Loopy Belief Propagation

SYSM 6304: Risk and Decision Analysis Lecture 5: Methods of Risk Analysis

Feedforward Neural Networks and Backpropagation

Evaluation of Machine Learning Techniques for Green Energy Prediction

3. The Junction Tree Algorithms

Solving Hybrid Markov Decision Processes

Bounded-width QBF is PSPACE-complete

Why? A central concept in Computer Science. Algorithms are ubiquitous.

Decision Trees and Networks

Chapter 7.30 Retrieving Medical Records Using Bayesian Networks

Lifted First-Order Belief Propagation

Question 2 Naïve Bayes (16 points)

Lecture 5 - CPA security, Pseudorandom functions

Model-based Synthesis. Tony O Hagan

Algebraic Computation Models. Algebraic Computation Models

Why Study NP- hardness. NP Hardness/Completeness Overview. P and NP. Scaling 9/3/13. Ron Parr CPS 570. NP hardness is not an AI topic

Persistent Binary Search Trees

Network File Storage with Graceful Performance Degradation

Removing Partial Inconsistency in Valuation- Based Systems*

ECLT5810 E-Commerce Data Mining Technique SAS Enterprise Miner -- Regression Model I. Regression Node

How To Solve A Minimum Set Covering Problem (Mcp)

IMPROVED HYBRID METHOD FOR CONSTRUCTING SMALL AND MEDIUM-SIZED BAYESIAN NETWORKS. Received September 2012; revised January 2013

Chapter 20: Data Analysis

Computational Geometry Lab: FEM BASIS FUNCTIONS FOR A TETRAHEDRON

Compression algorithm for Bayesian network modeling of binary systems

Satisfiability Checking

Determination of the normalization level of database schemas through equivalence classes of attributes

Chapter 4 - Lecture 1 Probability Density Functions and Cumul. Distribution Functions

Tracking Groups of Pedestrians in Video Sequences

Tetris is Hard: An Introduction to P vs NP

APP INVENTOR. Test Review

Integrating Benders decomposition within Constraint Programming

Scheduling Shop Scheduling. Tim Nieberg

A Web-based Bayesian Intelligent Tutoring System for Computer Programming

Categorical Data Visualization and Clustering Using Subjective Factors

Chapter 28. Bayesian Networks

Statistical Models in Data Mining

Chapter 15: Distributed Structures. Topology

The Basics of Graphical Models

A Bayesian Approach for on-line max auditing of Dynamic Statistical Databases

Gerry Hobbs, Department of Statistics, West Virginia University

Lecture 2: Universality

8. Machine Learning Applied Artificial Intelligence

University of Potsdam Faculty of Computer Science. Clause Learning in SAT Seminar Automatic Problem Solving WS 2005/06

Structured Learning and Prediction in Computer Vision. Contents

OHJ-2306 Introduction to Theoretical Computer Science, Fall

Incorporating Evidence in Bayesian networks with the Select Operator

1 Definitions. Supplementary Material for: Digraphs. Concept graphs

Chapter 13: Query Processing. Basic Steps in Query Processing

Bayesian Spam Detection

DETERMINING THE CONDITIONAL PROBABILITIES IN BAYESIAN NETWORKS

Diagonalization. Ahto Buldas. Lecture 3 of Complexity Theory October 8, Slides based on S.Aurora, B.Barak. Complexity Theory: A Modern Approach.

MINIMAL BOOKS OF RATIONALES

A Logical Approach to Factoring Belief Networks

Approximation Algorithms

Data Mining Practical Machine Learning Tools and Techniques

Probabilistic Networks An Introduction to Bayesian Networks and Influence Diagrams

Life of A Knowledge Base (KB)

Lecture 1: Course overview, circuits, and formulas

A Serial Partitioning Approach to Scaling Graph-Based Knowledge Discovery

Learning Sum-Product Networks with Direct and Indirect Variable Interactions

Probabilistic Evidential Reasoning with Symbolic Argumentation for Space Situation Awareness

Holland s GA Schema Theorem

Modeling and Performance Evaluation of Computer Systems Security Operation 1

Introduction to Algorithms. Part 3: P, NP Hard Problems

Bayesian Statistics: Indian Buffet Process

Topic: Greedy Approximations: Set Cover and Min Makespan Date: 1/30/06

Bayesian models of cognition

Study Manual. Probabilistic Reasoning. Silja Renooij. August 2015

CSC2420 Spring 2015: Lecture 3

A Bayesian Network Model for Diagnosis of Liver Disorders Agnieszka Onisko, M.S., 1,2 Marek J. Druzdzel, Ph.D., 1 and Hanna Wasyluk, M.D.,Ph.D.

Message-passing sequential detection of multiple change points in networks

Machine Learning.

Another Look at Sensitivity of Bayesian Networks to Imprecise Probabilities

TORA : Temporally Ordered Routing Algorithm

Case-Factor Diagrams for Structured Probabilistic Modeling

Bayesian Spam Filtering

Chapter 2 Introduction to Sequence Analysis for Human Behavior Understanding

Variational Mean Field for Graphical Models

Lecture 15 An Arithmetic Circuit Lowerbound and Flows in Graphs

WHERE DOES THE 10% CONDITION COME FROM?

A COMPARATIVE ASSESSMENT OF SUPERVISED DATA MINING TECHNIQUES FOR FRAUD PREVENTION

Development/Maintenance/Reuse: Software Evolution in Product Lines

Probability. Section 9. Probability. Probability of A = Number of outcomes for which A happens Total number of outcomes (sample space)

Invited Applications Paper

Tagging with Hidden Markov Models

Graphical Models for Recovering Probabilistic and Causal Queries from Missing Data

Tutorial 8. NP-Complete Problems

An Empirical Evaluation of Bayesian Networks Derived from Fault Trees

Service Performance Analysis and Improvement for a Ticket Queue with Balking Customers. Long Gao. joint work with Jihong Ou and Susan Xu

Compact Representations and Approximations for Compuation in Games

1 if 1 x 0 1 if 0 x 1

Transcription:

New Complexity Results for MAP in Bayesian Networks Dalle Molle Institute for Artificial Intelligence Switzerland IJCAI, 2011

Bayesian nets Directed acyclic graph (DAG) with nodes associated to (categorical) random variables; Collection of conditional probabilities p(x i Π i ) where Π i denotes the parents of X i in the graph (Π i may be empty); Every variable is conditionally independent of its non-descendants given its parents (Markov condition). New Complexity Results for MAP in Bayesian Networks Slide #1

Bayesian nets In other words, it is a compact way based on (in)dependence relations to represent a joint probability distribution. p(x 1,..., X n ) = i p(x i Π i ) New Complexity Results for MAP in Bayesian Networks Slide #2

Belief Updating BU: given a set of queried variables X and their states x, evidence variables E and their states e, compute p(x = x E = e). p(a d, e) = p(a, d, e) p(d, e) = b,c a,b,c p(a, b, c, d, e) p(a, b, c, d, e) New Complexity Results for MAP in Bayesian Networks Slide #3

Belief Updating - Decision version In terms of complexity, we can restrict ourselves to the computation of p(x = x, E = e). D-BU: given a rational α, a set of queried variables X and their states x, evidence variables E and their states e, decide whether p(x, e) > α. New Complexity Results for MAP in Bayesian Networks Slide #4

(Partial) Maximum a Posterior (MAP) MAP: given a set of queried variables X, evidence variables E and their states e, compute argmax x p(x e) = argmax x p(x, e). argmax a p(a, d, e) = argmax a p(a, b, c, d, e). b,c New Complexity Results for MAP in Bayesian Networks Slide #5

(Partial) Maximum a Posterior (MAP) - Decision version D-MAP: given a rational α, a set of queried variables X, evidence variables E and their states e, decide whether max x p(x, e) > α. New Complexity Results for MAP in Bayesian Networks Slide #6

Restricting Treewidth and Maximum cardinality MAP-z-w and D-MAP-z-w: same problems as before, but with two restrictions: z is a bound on the cardinality of any variable in the network, and w is a bound on the treewidth of the network. (The same definition can be used for the BU problem, which becomes BU-z-w and D-BU-z-w.) In order to express no bound, we use the symbol. E.g. D-MAP- - is the problem as defined earlier, and D-MAP- -w has a bound for treewidth, but not for for cardinality. New Complexity Results for MAP in Bayesian Networks Slide #7

Previous results Complexity of this problems had been studied before, including the case of bounded treewidth. D-BU- - is PP-complete, while D-BU- -w is in P. In fact, limiting the cardinality does not help: D-BU-2- is still PP-complete [Littman et al. 2001]. The functional versions are similar and discussed in [Roth 1996]. D-MAP- - is NP PP -complete, while D-MAP- -w is NP-complete [Park & Darwiche 2004]. MAP- -w is also shown not to be in Poly-APX [Park & Darwiche 2004]. (Unless P=NP) It is shown that there is no polynomial time approximation that can achieve a 2 bε -factor approximation, for 0 < ε < 1, b is the length of the input. New Complexity Results for MAP in Bayesian Networks Slide #8

... but cardinality has been neglected so far This paper presents new results for MAP that take cardinality into consideration. D-MAP-2-2 remains NP-complete (trick reduction from PARTITION). This includes binary polytrees. D-MAP- -1 remains NP-complete (reduction from MAX-2-SAT using a naive-like structure) and D-MAP-5-1 is NP-complete too (reduction from PARTITION using an HMM-like structure). This includes even simple trees. It is NP-hard to approximate MAP- -1 to any factor 2 bε (the construction comes from the naive-like structure, and uses similar arguments as in [Park & Darwiche 2004]). New Complexity Results for MAP in Bayesian Networks Slide #9

Decision problems P NP DMPE- - PP NP PP DMPE- -w DBU- -w DMPE-2- DMAP- -2 DMAP- -1 DBU-2- DBU- - DMAP-2- DMAP- - DMAP-2-2 DMAP-5-1 New Complexity Results for MAP in Bayesian Networks Slide #10

... and there is (some) hope... MAP-z-w is hard, but has a FPTAS! We develop a Fully Polynomial-time Approximation Scheme for MAP when both treewidth and cardinality are bounded. The idea is to compute all possible candidates and propagate them as in a BU inference, but keeping the number of candidates bounded by a polynomial in the length of the input (following ideas from [Papadimitriou and Yannakakis, 2000]). Previous inapproximability results are not contradicted: they had used variables with high cardinality. New Complexity Results for MAP in Bayesian Networks Slide #11

Functional problems FP exp-apx poly-apx FPTAS NPO? BU- -w MAP- -1 MPE-2- MAP-2- MPE- -w MAP-z-w MAP- -2 MPE- - MAP- - MAP- -w New Complexity Results for MAP in Bayesian Networks Slide #12

Conclusions This paper targets on understanding better the computational complexity of MAP. The problem is shown to remain hard in binary polytrees and trees with bounded cardinality. The problem is shown to be not approximable even in trees (without cardinality restrictions). An FPTAS is devised when both treewidth and cardinality are bounded. New Complexity Results for MAP in Bayesian Networks Slide #13

Thanks Thank you for your attention. Further questions: cassiopc@acm.org Work partially supported by Project Computational Life Sciences - Ticino in Rete, Switzerland. Grant from the Swiss NSF n. 200020 134759/1. New Complexity Results for MAP in Bayesian Networks Slide #14