dtproblog
play

DTProbLog A Decision-Theoretic Probabilistic Prolog Guy Van den - PowerPoint PPT Presentation

Mar 23, 2023 •368 likes •812 views

DTProbLog A Decision-Theoretic Probabilistic Prolog Guy Van den Broeck Ingo Thon Martijn van Otterlo Luc De Raedt Motivation Many real-world decision problems are relational and probabilistic Wildfire control Who to vaccinate for

  1. DTProbLog A Decision-Theoretic Probabilistic Prolog Guy Van den Broeck Ingo Thon Martijn van Otterlo Luc De Raedt

  2. Motivation ● Many real-world decision problems are relational and probabilistic ● Wildfire control ● Who to vaccinate for swine flu ● Viral marketing ● ... 07/14/10 DTProbLog 2

  3. Motivation ● Combination of relations , uncertainty and decision theory largely ignored some exceptions are MLDNs, DTLPs and FOMDPs ● Relations with uncertainty = S tatistical R elational L earning ● ProbLog is a simple probabilistic Prolog ● DTProbLog is a decision-theoretic ProbLog 07/14/10 DTProbLog 3

  4. Outline ● DTProbLog: the Language ● DTProbLog: the Algorithms ● Exact Solution Algorithm ● Approximate Solution Algorithms ● Experiments: Viral Marketing ● Related Work & Conclusions 07/14/10 DTProbLog 4

  5. DTProbLog: the Language 07/14/10 DTProbLog 5

  6. Maggie Homer Lenny Marge Moe Bart Lisa Apu Seymour Ralph 07/14/10 DTProbLog 6

  7. Maggie Homer Lenny Marge 1 Moe Bart Lisa Apu Seymour Ralph 07/14/10 DTProbLog 7

  8. Maggie Homer Lenny Marge 1 Moe 0.3 0.3 Bart Lisa 0.3 Apu Seymour Ralph 07/14/10 DTProbLog 8

  9. Maggie 0.17 Homer Lenny Marge 1 0.14 Moe 0.39 0.34 Bart Lisa 0.41 0.25 Apu 0.17 Seymour Ralph 0.25 0.21 07/14/10 DTProbLog 9

  10. ProbLog Probabilistic Facts 0.3 :: buy_trust(_,_). Background Knowledge buys(X) :- trusts(X,Y), buys(Y), buy_trust(X,Y). Maggie 0.17 Homer Lenny Marge 1 0.14 Moe 0.39 0.34 Bart Lisa 0.41 0.25 Apu 0.17 Seymour Ralph 0.25 0.21 07/14/10 DTProbLog 10

  11. Maggie Homer Lenny Marge Moe Bart Lisa Apu Seymour Ralph 07/14/10 DTProbLog 11

  12. Maggie Homer Lenny Marge 0.20 Moe Bart Lisa Apu Seymour Ralph 07/14/10 DTProbLog 12

  13. Maggie 0.03 Homer Lenny Marge 0.20 0.03 Moe 0.08 0.07 Bart Lisa 0.08 0.05 Apu 0.03 Seymour Ralph 0.05 0.04 07/14/10 DTProbLog 13

  14. ProbLog Probabilistic Facts 0.3 :: buy_trust(_,_). 0.2 :: buy_marketing(_). Background Knowledge buys(X) :- buys(X) :- trusts(X,Y), marketed(X), buys(Y), buy_marketing(X). buy_trust(X,Y). Maggie 0.03 Homer Lenny Marge 0.20 0.03 Moe 0.08 0.07 Bart Lisa 0.08 0.05 Apu 0.03 Seymour Ralph 0.05 0.04 07/14/10 DTProbLog 14

  15. Maggie Homer Lenny Marge Moe Bart Lisa Apu Seymour Ralph 07/14/10 DTProbLog 15

  16. Maggie 0.05 Homer Lenny Marge 0.25 0.10 Moe 0.11 0.25 Bart Lisa 0.12 0.08 Apu 0.11 Seymour Ralph 0.12 0.08 07/14/10 DTProbLog 16

  17. +$5 Which strategy gives the maximum -$3 expected utility ? ? ? Maggie ? ? ? ? ? ? Homer Lenny ? ? Marge Moe ? ? ? ? Bart Lisa ? ? ? ? Apu ? ? Seymour Ralph 07/14/10 DTProbLog 17

  18. +$5 2 -$3 1 . 1 $ Maggie 0.07 Homer Lenny Marge 0.12 0.06 Moe 0.13 0.12 Bart Lisa 0.27 0.13 Apu 0.11 Seymour Ralph 0.27 0.16 07/14/10 DTProbLog 18

  19. Probabilistic Facts DT ProbLog ... Background Knowledge ... Decisions ? :: marketed(P) :- person(P). 2 1 Utility Facts . 1 $ buys(P) => 5 :- person(P). marketed(P) => -3 :- person(P). Maggie 0.07 Homer Lenny Marge 0.12 0.06 Moe 0.13 0.12 Bart Lisa 0.27 0.13 Apu 0.11 Seymour Ralph 0.27 0.16 07/14/10 DTProbLog 19

  20. DTProbLog: the Algorithms 07/14/10 DTProbLog 20

  21. DTProbLog: the Algorithms ● DTProbLog solves decision problems in complex relational and uncertain environments. ● Exact solution algorithm ● Extends ProbLog's BDD-based inference ● Efficient datastructures: BDD and ADD ● Approximate algorithms ● Local search ● K-best proofs 07/14/10 DTProbLog 21

  22. Example: Dressing for unpredictable weather Decision Facts Probabilistic Facts ? :: umbrella. 0.3 :: rainy . ? :: raincoat. 0.5 :: windy . Background Knowledge dry :- rainy, umbrella, not(broken_umbrella). dry :- rainy, raincoat. dry :- not(rainy). broken_umbrella :- umbrella, rainy, windy . Utility Facts umbrella => -2. dry => 60. raincoat => -20. broken_umbrella => -40. 07/14/10 DTProbLog 22

  23. Exact Solution Algorithm Find all proofs for each utility attribute (Prolog) 1 Binary decision diagrams from the proofs 2 Algebraic decision diagrams for the probability 3 of each attribute Algebraic decision diagrams for the utility of 4 each attribute Algebraic decision diagram for the total utility 5 07/14/10 DTProbLog 23

  24. Find all proofs for each utility attribute (Prolog) 1 Binary decision diagrams from the proofs 2 ● Leafs indicate attribute true or false ● Nodes are probabilistic facts or decisions true false dry => 60. umbrella => -2. broken_umbrella => -40. raincoat => -20. 07/14/10 DTProbLog 24

  25. Algebraic decision diagrams for the probability 3 of each attribute ● Probabilistic facts are marginalized out ● Nodes are decisions only ● Leafs are probabilities dry => 60. umbrella => -2. broken_umbrella => -40. raincoat => -20. 07/14/10 DTProbLog 25

  26. Algebraic decision diagrams for the utility of 4 each attribute ● Leafs are expected utilities dry => 60. umbrella => -2. broken_umbrella => -40. raincoat => -20. 07/14/10 DTProbLog 26

  27. Algebraic decision diagram for the total utility 5 ∑ + sound pruning (ADD not built entirely in memory) 07/14/10 DTProbLog 27

  28. Approximate Solution: Local Search Find all proofs for each utility attribute (Prolog) 1 Binary decision diagrams from the proofs 2 Algebraic decision diagrams for the probability 3 of each attribute Algebraic decision diagrams for the utility of 4 each attribute Algebraic decision diagram for the total utility 5 07/14/10 DTProbLog 28

  29. Approximate Solution: Local Search Find all proofs for each utility attribute (Prolog) 1 Binary decision diagrams from the proofs 2 Algebraic decision diagrams for the probability 3 of each attribute Algebraic decision diagrams for the utility of 4 each attribute Algebraic decision diagram for the total utility 5 Greedy hillclimber search 3 07/14/10 DTProbLog 29

  30. Approximate Solution: K-best Proofs Find all proofs for each utility attribute (Prolog) 1 Binary decision diagrams from the proofs 2 Algebraic decision diagrams for the probability 3 of each attribute Algebraic decision diagrams for the utility of 4 each attribute Algebraic decision diagram for the total utility 5 07/14/10 DTProbLog 30

  31. Approximate Solution: K-best Proofs Find all proofs for each utility attribute (Prolog) 1 Find the k most likely proofs for each utility 1 attribute Binary decision diagrams from the proofs 2 Algebraic decision diagrams for the probability 3 of each attribute Algebraic decision diagrams for the utility of 4 each attribute Algebraic decision diagram for the total utility 5 07/14/10 DTProbLog 31

  32. Experiments: Viral Marketing 07/14/10 DTProbLog 32

  33. Experiments: Viral Marketing ● Synthetic dataset Random power law graphs of increasing size ● Real-world Epinions dataset [Domingos02] 07/14/10 DTProbLog 33

  34. (Q1) Does the exact solution algorithm perform better than naively calculating the utility of all strategies? 07/14/10 DTProbLog 34

  35. (Q2) What is the difference in runtime and solution quality between exact and local search? 07/14/10 DTProbLog 35

  36. (Q3) What is the difference in runtime and solution quality between for different values of k in k-best . 07/14/10 DTProbLog 36

  37. (Q4) Do the algorithms scale ? Epinions social network ● 75,000 people ● 500,000 trust relations YES ● Local search ● 17-best proofs ● Solved in 16 hours 07/14/10 DTProbLog 37

  38. Related Work & Conclusions 07/14/10 DTProbLog 38

  39. Related Work Representation Solution Evaluation Global Local Exact Approximate Relational Probabilities optimum optimum inference inference Influence Diagrams MLDNs [Nath] ICL [Poole] DTLPs [Chen] DTProbLog 07/14/10 DTProbLog 39

  40. Summary ● DTProbLog, the programming language ● Probabilistic Prolog ● Decisions ● Utilities: rewards or costs attached to goals ● Solution algorithms ● Exactly ● Approximately ● Experiments ● Effective ● Scale well 07/14/10 DTProbLog 40

  41. Ongoing and Future Work ● Sequential decision problems ● Easy to represented in DTProbLog ● Bad fit for solution algorithms ● Solvers ● Integer linear programming ● Bounded approximation ● Monte-Carlo ● Lifting (many BDDs have same structure) ● Learning DTProbLog programs ● Inverse reinforcement learning 07/14/10 DTProbLog 41

  42. Thank You! 07/14/10 DTProbLog 42

  43. Viral Marketing Decisions ? :: marketed(P) :- person(P). Probabilistic Facts 0.3 :: buy_trust(_,_). 0.2 :: buy_marketing(_). Background Knowledge buys(X) :- buys(X) :- trusts(X,Y), marketed(X), buys(Y), buy_marketing(X). buy_trust(X,Y). Utility Facts buys(P) => 5 :- person(P). marketed(P) => -3 :- person(P). 07/14/10 DTProbLog 43

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Announcements Solutions to Problem Set 3 are posted Problem Set 4 is posted, It will be graded

Announcements Solutions to Problem Set 3 are posted Problem Set 4 is posted, It will be graded

Announcements Solutions to Problem Set 3 are posted Problem Set 4 is posted, It will be graded and is due a week from Friday You already know everything you need to work on Problem Set 4 Professor Miller will be filling in for me in

799 views • 45 slides
even more ways to generate self log missed opportunities data lost forever different modes of

even more ways to generate self log missed opportunities data lost forever different modes of

even more ways to generate self log missed opportunities data lost forever different modes of self logging keep calm and thank google what i did cant you tell i was falling in love? how i did it timestamp | sender | message chat

307 views • 28 slides
CPSC 111 Introduction to Computation Introduction to Computation October 1 st , 2009 Based on

CPSC 111 Introduction to Computation Introduction to Computation October 1 st , 2009 Based on

CPSC 111 Introduction to Computation Introduction to Computation October 1 st , 2009 Based on slides by Eiselt, Carter, Murphy, Pottinger Administrative Stuff Administrative Stuff If you have flu symptoms, there's a good chance that it really

619 views • 50 slides
Modelling in fm uenza-like illness using online search Vasileios Lampos Computer Science , UCL

Modelling in fm uenza-like illness using online search Vasileios Lampos Computer Science , UCL

Modelling in fm uenza-like illness using online search Vasileios Lampos Computer Science , UCL @lampos lampos.net www Mapping online search to fm u estimates 12 10 ILI percentage 8 6 4 2 0 2004 2005 2006 2007 2008 Year Why estimate

638 views • 36 slides
GEOVISUALIZATION IN NEOGEOGRAPHY? TANMOY DAS MENNO-JAN KRAAK CONTENTS Introduction

GEOVISUALIZATION IN NEOGEOGRAPHY? TANMOY DAS MENNO-JAN KRAAK CONTENTS Introduction

GEOVISUALIZATION IN NEOGEOGRAPHY? TANMOY DAS MENNO-JAN KRAAK CONTENTS Introduction Terminology Potential Applications Problems Methods Conclusions INTRODUCTION Applications of Web 2.0 Photo & Video Blogs, Tweets

270 views • 11 slides
San Mateo County Fire In Cooperation with Mission Statement CAL FIRE serves and safeguards the

San Mateo County Fire In Cooperation with Mission Statement CAL FIRE serves and safeguards the

San Mateo County Fire In Cooperation with Mission Statement CAL FIRE serves and safeguards the people and protects the property and resources of California. Shared Vision 2025 Ensure that our neighborhoods are safe and provide residents

295 views • 27 slides
Syndromic Surveillance Predicting and monitoring influenza-like illness using telephone and

Syndromic Surveillance Predicting and monitoring influenza-like illness using telephone and

Syndromic Surveillance Predicting and monitoring influenza-like illness using telephone and digital data Work by Mica Hartley Presented by Michael Araco Outline Influenza a healthcare problem How can telephone and digital data help?

397 views • 14 slides
Inclusive Growth and the Informal Waste Sector in Africa: Exploring the Challenges and

Inclusive Growth and the Informal Waste Sector in Africa: Exploring the Challenges and

Inclusive Growth and the Informal Waste Sector in Africa: Exploring the Challenges and Opportunities Martin Medina Contact: martin.medina-martinez.grd.genr@aya.yale.edu Waste Management in Africa One of the Most Pressing Urban Environmental

928 views • 43 slides
1 2 3 21 st CC Confident Person, Concerned Citizen, Seld-Directed Learner, Acitve

1 2 3 21 st CC Confident Person, Concerned Citizen, Seld-Directed Learner, Acitve

1 2 3 21 st CC Confident Person, Concerned Citizen, Seld-Directed Learner, Acitve Contributor M otto kept though there was talk in the previous leadership to change to the Best is Y et to Be. The existing motto is kept as the school

1.29k views • 77 slides
Turning Your Cybersecurity Toddlers Into Warriors! Simple lessons to fill the knowledge gap

Turning Your Cybersecurity Toddlers Into Warriors! Simple lessons to fill the knowledge gap

Turning Your Cybersecurity Toddlers Into Warriors! Simple lessons to fill the knowledge gap within your staff Shira Shamban Dome9 Security @shambanIT 0 1 2 3 4 5 You and your staff Are NOT going to keep up with Technology

483 views • 31 slides
Exploring Views for Goal-oriented Requirements Comprehension Lyrene Silva, Ana Moreira, Joo

Exploring Views for Goal-oriented Requirements Comprehension Lyrene Silva, Ana Moreira, Joo

Exploring Views for Goal-oriented Requirements Comprehension Lyrene Silva, Ana Moreira, Joo Arajo, Catarina Gralha , Miguel Goulo and Vasco Amaral Universidade Federal do Rio Grande do Norte (Brazil), Universidade NOVA de

446 views • 22 slides
INF5470 Spring 2010 Philipp Hfliger Lecture 2: Neurophysiology in a Nutshell Content

INF5470 Spring 2010 Philipp Hfliger Lecture 2: Neurophysiology in a Nutshell Content

INF5470 Spring 2010 Philipp Hfliger Lecture 2: Neurophysiology in a Nutshell Content Overview Methods Known Properties Lecture 2: Neurophysiology in a Nutshell 2 Content Overview Methods Known Properties Lecture 2: Neurophysiology

685 views • 29 slides
Digital Medicine I: Introduction to Programming Introduction to the Course Autumn 2019

Digital Medicine I: Introduction to Programming Introduction to the Course Autumn 2019

Hans-Joachim Bckenhauer and Dennis Komm Digital Medicine I: Introduction to Programming Introduction to the Course Autumn 2019 October 3, 2019 Welcome to the Course Material Lecture website https://courses.ite.inf.ethz.ch/digiMed19

1.4k views • 100 slides
Ann Arbor VA Credentialing Academic Year 2020 - 2021 VA Trainee On-boarding 1. Trainee Guide and

Ann Arbor VA Credentialing Academic Year 2020 - 2021 VA Trainee On-boarding 1. Trainee Guide and

Ann Arbor VA Credentialing Academic Year 2020 - 2021 VA Trainee On-boarding 1. Trainee Guide and Requirements Talent Management System (TMS) Fingerprints How to complete and submit VA Paperwork 2. Trainee Application New to the VA

468 views • 16 slides
DIADS: Addressing the My-Problem- or-Yours Syndrome with Integrated SAN and Database

DIADS: Addressing the My-Problem- or-Yours Syndrome with Integrated SAN and Database

DIADS: Addressing the My-Problem- or-Yours Syndrome with Integrated SAN and Database Diagnosis Nedyalko Borisov Duke University Shivnath Babu, Duke Sandeep Uttamchandani, IBM Ramani Routray, IBM Aameek Singh, IBM Current State

482 views • 36 slides
SSD Fail ilures in in Datacenters: What? When? And Why? Iyswarya Narayanan, Di Wang, Myeongjae

SSD Fail ilures in in Datacenters: What? When? And Why? Iyswarya Narayanan, Di Wang, Myeongjae

SSD Fail ilures in in Datacenters: What? When? And Why? Iyswarya Narayanan, Di Wang, Myeongjae Jeon, Bikash Sharma, Laura Caulfield, Anand Sivasubramaniam, Ben Cutler, Jie Liu, Badriddine Khessib, Kushagra Vaid The 9 th ACM Systems And Storage

675 views • 39 slides
Public Believes Ebola is Likely to Spread from Symptomatic Patients by Multiple Routes % saying

Public Believes Ebola is Likely to Spread from Symptomatic Patients by Multiple Routes % saying

Public Believes Ebola is Likely to Spread from Symptomatic Patients by Multiple Routes % saying Ebola is likely to spread from a symptomatic person to other people (81%) 49% 32% % saying how likely a person is to get Ebola if they Come into

71 views • 3 slides
The CARES Act & Return to Work Overview & Major Provisions for Individuals $2 Trillion

The CARES Act & Return to Work Overview & Major Provisions for Individuals $2 Trillion

The CARES Act & Return to Work Overview & Major Provisions for Individuals $2 Trillion Package Through the Federal Reserve potential for additional $4 trillion in support for the economy Referred to as the Phase 3 Bill

447 views • 8 slides
SnappyData: Apache Spark Meets Embedded In-Memory Database Masaki Yamakawa UL Systems, Inc.

SnappyData: Apache Spark Meets Embedded In-Memory Database Masaki Yamakawa UL Systems, Inc.

SnappyData: Apache Spark Meets Embedded In-Memory Database Masaki Yamakawa UL Systems, Inc. About me Masaki Yamakawa UL Systems, Inc. Managing Consultant { Sector Financial Skills [ Distributed system ,

674 views • 51 slides
Modelization of membrane potentials and information transmission in large systems of neurons

Modelization of membrane potentials and information transmission in large systems of neurons

introduction membrane potential as a (jump) diffusion process Poisson spike trains information transmission in large systems of neurons statistical inference, Modelization of membrane potentials and information transmission in large systems of

507 views • 28 slides
Neural Networks - Deep Learning Artificial Intelligence @ Allegheny College Janyl Jumadinova

Neural Networks - Deep Learning Artificial Intelligence @ Allegheny College Janyl Jumadinova

Neural Networks - Deep Learning Artificial Intelligence @ Allegheny College Janyl Jumadinova March 4-6, 2020 Credit: Google Workshop Janyl Jumadinova March 4-6, 2020 1 / 35 Neural Networks - Deep Learning Neural Networks Janyl Jumadinova

709 views • 48 slides
Board AND COMPUTER, PLEASE LET ADAM KNOW WHAT PHONE NUMBER YOU USED TO DIAL IN SO YOU CAN BE

Board AND COMPUTER, PLEASE LET ADAM KNOW WHAT PHONE NUMBER YOU USED TO DIAL IN SO YOU CAN BE

ENTER YOUR NAME WHEN YOU ENTER THE CONFERENCE ROOM. ALL VIRTUAL ATTENDEES ARE MUTED BY THE MEETING HOST FROM THE BEGINNING. IF YOU ARE A BOARD MEMBER JOINING BY BOTH PHONE Board AND COMPUTER, PLEASE LET ADAM KNOW WHAT PHONE NUMBER

1.11k views • 85 slides
Introduction to Artificial Intelligence Neural Networks - Deep Learning for NLP Janyl Jumadinova

Introduction to Artificial Intelligence Neural Networks - Deep Learning for NLP Janyl Jumadinova

Introduction to Artificial Intelligence Neural Networks - Deep Learning for NLP Janyl Jumadinova November 21, 2016 Neural Networks 2/20 Neural Networks 3/20 Neural Networks Neural computing requires a number of neurons , to be connected

813 views • 21 slides
Connectomics! Neurons in the brain illustration Image credit: Benedict Campbell. Welcome

Connectomics! Neurons in the brain illustration Image credit: Benedict Campbell. Welcome

Connectomics! Neurons in the brain illustration Image credit: Benedict Campbell. Welcome Images Neural Structure Reconstruction from Fluorescence Imaging of Neural Activity Karan Singh Pankaj Gupta Objectives Communicate research

887 views • 25 slides

More recommend