State Estimation: Particle Filter

Size: px
Start display at page:

Download "State Estimation: Particle Filter"

Transcription

1 State Estimation: Particle Filter Daniel Seliger HAUPT-/ BACHELOR- SEMINAR ADAPTIVE SYSTEME PST PROF. DR. WIRSING 14. JUNI 2009 VORNAME NAME

2 Overview 1. Repitition: Bayesian Filtering 2. Particle Filter 4. Summary 5. Sources Daniel Seliger 2

3 Important Note Note: The terms probability, probability density, probability distribution are mixed up in this presentation and may not be mathematically correct. However, there's not a lot of maths and so the difference shouldn't be important to understand the topic. S. Thrun, W. Burgard, D. Fox also did this in Probabilistic Robotics, The MIT Press Daniel Seliger 3

4 Repitition Bayesian Filters Control Actions u Robot Environment Sensor Measurements z Initial state: Posterior distribution: State transition probability: How does state change over time (e.g. as an effect of a control action)? Measurement probability: x t x 0 p(x t u t, z t ) p(x t x t 1,u t ) p(z t x t ) Assumed state,how probable is measurement? z t Daniel Seliger 4

5 Repitition Bayesian Filters Initialization x 0 Correction p(z t x t ) Prediction p(x t x t 1,u t ) Goal: Extract estimate (e.g. mean of posterior) Update the posterior distribution over time, taking into account the controls and measurements, in a way that the extracted estimate converges against the real state Daniel Seliger 5

6 Parametric vs Non-Parametric Parametric filters: Continous state space Describe probabilities in parametric terms: Advantage: Best possible accuracy Disadvantage: Only simple distributions Example: Kalman filter Abb.1: Dichtefunktion der Standardnormaverteilung Daniel Seliger 6

7 Parametric vs Non-Parametric Non-Parametric filters: Break up continous state space into discrete state space Describe probabilities by a set of samples Advantage: Ability to describe complex probability distributions Disadvantage: Need to trade off accuracy against efficiency Example: Histogram Filter, or... Abb. 2: Histogram representation of a continuous random variable Daniel Seliger 7

8 Particle Filter Daniel Seliger 8

9 Many different names:... Particle Filter Sequential Monte-Carlo-Method CONDENSATION algorithm Interacting particle approximations Bootstrap filtering Survival of the fittest Daniel Seliger 9

10 Particle: Concrete belief of the state, combined with a Weight w that indicates the certainty that the belief is true Daniel Seliger 10

11 Particle: Concrete belief of the state, combined with a Weight w that indicates the certainty that the belief is true Set of Particles: approximation of the posterior distribution from which an estimation of the real state can be extracted Daniel Seliger 11

12 Sampling: draw a number of samples (particles) acchording to a given probability Abb. 3: samples drawn from a Gaussian random variable Daniel Seliger 12

13 Functional Principle of Particle Filter: Initialisation Prediction Resampling Importance Sampling Estimation Daniel Seliger 13

14 Example: Mobile Robot Localisation Abb. 4 x (state space) Daniel Seliger 14

15 Example: Mobile Robot Localisation Abb. 4 x (state space) Robot knows the room, but he doesn't know where he is Robot has a sensor that indicates door / no door Goal: locate Robot! Daniel Seliger 15

16 Functional Principle of Particle Filter: Initialisation Prediction Resampling Importance Sampling Estimation Daniel Seliger 16

17 Initialisation: Robot doesn't have a clue where he is yet, so Randomly draw N particles over the whole state space All particles have the same weight w Abb. 5: Particle Filter Initialisation x Daniel Seliger 17

18 Functional Principle of Particle Filter: Initialisation Prediction Resampling Importance Sampling Estimation Daniel Seliger 18

19 Importance Sampling: After the sensor has made a measurement ( door), weigh the particles according to the measurement probability Particles (=beliefs) with a high probability get a high weight and vice versa Normalize the weights Abb. 6: Importance Sampling p(z t x t ) w x Daniel Seliger 19 x

20 Functional Principle of Particle Filter: Initialisation Prediction Resampling Importance Sampling Estimation Daniel Seliger 20

21 Resampling: The set of weighed particles describes a probability distribution Pick N particles, with replacement, according to that distribution Particles with little weight are eliminated, while others are picked more often Particles concentrate in areas of higher probability There are still N particles Some particles represent the same belief resampling x Abb. 7: Particle Resampling Daniel Seliger 21

22 Functional Principle of Particle Filter: Initialisation Prediction Resampling Importance Sampling Estimation Daniel Seliger 22

23 Prediction: The Robot has moved (control action) Move each particle according to the state transition probability p(x t x t 1,u t ) Abb. 8: Prediction x Increase of granularity! Daniel Seliger 23

24 Resampling and Prediction: w x w Abb. 9: Resampling and Prediction x Daniel Seliger 24

25 Evolution of particles after one loop: Before: w x After: w Abb. 10: Particle evolution after one loop x Particles (beliefs) are concentrated on more probable states Daniel Seliger 25

26 Evolution of particles after two loops: Before: w x After: w Abb. 11: Particle evolution after two loops x The distribution of the particles tends to become a more precise posterior distribution from which we can extract an accurate estimate Daniel Seliger 26

27 Implementation in pseudo-code: Last set of particles Auxilliary set Resulting set Prediction Resampling Importance Sampling Estimation Abb. 12: Particle Filter implementation Daniel Seliger 27

28 Implementation in pseudo-code: Prediction Resampling Importance Sampling Estimation Abb. 12: Particle Filter implementation Daniel Seliger 28

29 Implementation in pseudo-code: Prediction Resampling Importance Sampling Estimation Abb. 12: Particle Filter implementation Daniel Seliger 29

30 Implementation in pseudo-code: Prediction Resampling Importance Sampling Estimation Abb. 12: Particle Filter implementation Daniel Seliger 30

31 Robot Localisation in 3-dimensional state space: Abb. 13: Mobile robot localisation via sonar sensors and Particle Filter Daniel Seliger 31

32 (Dis-)Advantages of PF Particle Filter: Advantages: No Limitation for the complexity of distributions Continous and discrete state spaces Adaptive in the use of computational resources Disadvantages: The most efficient number of particles cannot be calculated Distributions are only approximated which leads to calculation errors Daniel Seliger 32

33 Conclusion What is the best filter for state estimation? There is none! Each filter has advantages over the others The choice of filter depends on the concrete system of which you want to estimate the state In many applications, several filters are used together, e.g. Kalman Filter and Particle Filter Daniel Seliger 33

34 Summary Summary: Bayesian Filters are used to estimate states Real state cannot be retrieved, but approximated by a probability distribution over time Prediction / Correction cycle Parametric and Non-Parametric filters Parametric filters are more accurate but cannot handle complex distributions Non-Parametric filters can deal with very complex distributions but are less accurate because of approximation Daniel Seliger 34

35 Summary Particle Filter: Describes posterior by a set of weighted samples Each sample represents a concrete belief Importance Sampling: weigh each particle according to the measurement distribution Resampling: sample particles new, according to their weight force particles to concentrate in regions of higher interest Prediction: move each particle according to the state transition probability increase granularity Conclusion: Each filter has different advantages and disadvantages Choice of filter depends on field of application Daniel Seliger 35

36 Sources Literature: S. Thrun, W. Burgard, D. Fox: Probabilistic Robotics, The MIT Press 2005 AI-Videos: Robotics-Videos: Drew Bagnell: Statistical Techniques in Robotics Lecture #4, 2008: f11/notes/f08/lecture4/16831_lecture04.mdesnoye.pdf Miodrag Bolic: Theory and Implementation of Particle Filters, University of Ottawa Daniel Seliger 36

37 Sources Pieter Abbeel: Particle Filters, Michael Pfeiffer: A brief introduction to Particle Filters, TU Graz 2004, Wikipedia: Particle Filter, Images: Abb. 1: Dichtefunktion der Standardnormalverteilung, Wikipedia: Normalverteilung, Daniel Seliger 37

38 Sources Abb. 2 - Abb. 6: S. Thrun, W. Burgard, D. Fox: Probabilistic Robotics, The MIT Press 2005 Abb. 7, Abb. 8: Miodrag Bolic: Theory and Implementation of Particle Filters, University of Ottawa 2004 Abb. 9 - Abb. 12: S. Thrun, W. Burgard, D. Fox: Probabilistic Robotics, The MIT Press 2005 Abb. 13: D. Fox, Daniel Seliger 38

A Vision-based Affective Computing System. Jieyu Zhao Ningbo University, China

A Vision-based Affective Computing System. Jieyu Zhao Ningbo University, China A Vision-based Affective Computing System Jieyu Zhao Ningbo University, China Outline Affective Computing A Dynamic 3D Morphable Model Facial Expression Recognition Probabilistic Graphical Models Some

More information

AClass: A Simple, Online Probabilistic Classifier. Vikash K. Mansinghka Computational Cognitive Science Group MIT BCS/CSAIL

AClass: A Simple, Online Probabilistic Classifier. Vikash K. Mansinghka Computational Cognitive Science Group MIT BCS/CSAIL AClass: A Simple, Online Probabilistic Classifier Vikash K. Mansinghka Computational Cognitive Science Group MIT BCS/CSAIL AClass: A Simple, Online Probabilistic Classifier or How I learned to stop worrying

More information

Mathematical Microbiologists: Why we have to return to our square roots to uncover uncertainty (of measurement) in Quantitative PCR (qpcr)

Mathematical Microbiologists: Why we have to return to our square roots to uncover uncertainty (of measurement) in Quantitative PCR (qpcr) Mathematical Microbiologists: Why we have to return to our square roots to uncover uncertainty (of measurement) in Quantitative PCR (qpcr) J. Ian Stuart George Zahariadis Marina Salvadori Objectives 1.

More information

Bayesian Statistics Estimation of a Single Mean and Variance MCMC Diagnostics and Missing Data

Bayesian Statistics Estimation of a Single Mean and Variance MCMC Diagnostics and Missing Data Bayesian Statistics Estimation of a Single Mean and Variance MCMC Diagnostics and Missing Data Michael Anderson, PhD Hélène Carabin, DVM, PhD Department of Biostatistics and Epidemiology The University

More information

Seminar Thesis: Efficient Planning under Uncertainty with Macro-actions

Seminar Thesis: Efficient Planning under Uncertainty with Macro-actions Seminar Thesis: Efficient Planning under Uncertainty with Macro-actions Ragnar Mogk Department of Computer Science Technische Universität Darmstadt ragnar.mogk@stud.tu-darmstadt.de 1 Introduction This

More information

Exploring the Influence of Particle Filter Parameters on Order Effects in Causal Learning

Exploring the Influence of Particle Filter Parameters on Order Effects in Causal Learning Exploring the Influence of Particle Filter Parameters on Order Effects in Causal Learning Joshua T. Abbott (joshua.abbott@berkeley.edu) Thomas L. Griffiths (tom griffiths@berkeley.edu) Department of Psychology,

More information

CS343: Artificial Intelligence

CS343: Artificial Intelligence CS343: Artificial Intelligence Introduction: Part 2 Prof. Scott Niekum University of Texas at Austin [Based on slides created by Dan Klein and Pieter Abbeel for CS188 Intro to AI at UC Berkeley. All materials

More information

Introduction to Computational Neuroscience

Introduction to Computational Neuroscience Introduction to Computational Neuroscience Lecture 10: Brain-Computer Interfaces Ilya Kuzovkin So Far Stimulus So Far So Far Stimulus What are the neuroimaging techniques you know about? Stimulus So Far

More information

MSc Neuroimaging for Clinical & Cognitive Neuroscience

MSc Neuroimaging for Clinical & Cognitive Neuroscience MSc Neuroimaging for Clinical & Cognitive Neuroscience School of Psychological Sciences Faculty of Medical & Human Sciences Module Information *Please note that this is a sample guide to modules. The exact

More information

Reach and grasp by people with tetraplegia using a neurally controlled robotic arm

Reach and grasp by people with tetraplegia using a neurally controlled robotic arm Leigh R. Hochberg et al. Reach and grasp by people with tetraplegia using a neurally controlled robotic arm Nature, 17 May 2012 Paper overview Ilya Kuzovkin 11 April 2014, Tartu etc How it works?

More information

Spatial Cognition for Mobile Robots: A Hierarchical Probabilistic Concept- Oriented Representation of Space

Spatial Cognition for Mobile Robots: A Hierarchical Probabilistic Concept- Oriented Representation of Space Spatial Cognition for Mobile Robots: A Hierarchical Probabilistic Concept- Oriented Representation of Space Shrihari Vasudevan Advisor: Prof. Dr. Roland Siegwart Autonomous Systems Lab, ETH Zurich, Switzerland.

More information

Dynamic Control Models as State Abstractions

Dynamic Control Models as State Abstractions University of Massachusetts Amherst From the SelectedWorks of Roderic Grupen 998 Dynamic Control Models as State Abstractions Jefferson A. Coelho Roderic Grupen, University of Massachusetts - Amherst Available

More information

Sequential Decision Making

Sequential Decision Making Sequential Decision Making Sequential decisions Many (most) real world problems cannot be solved with a single action. Need a longer horizon Ex: Sequential decision problems We start at START and want

More information

Introduction. Patrick Breheny. January 10. The meaning of probability The Bayesian approach Preview of MCMC methods

Introduction. Patrick Breheny. January 10. The meaning of probability The Bayesian approach Preview of MCMC methods Introduction Patrick Breheny January 10 Patrick Breheny BST 701: Bayesian Modeling in Biostatistics 1/25 Introductory example: Jane s twins Suppose you have a friend named Jane who is pregnant with twins

More information

ENHANCED VEHICLE IDENTIFICATION UTILIZING SENSOR FUSION AND STATISTICAL ALGORITHMS

ENHANCED VEHICLE IDENTIFICATION UTILIZING SENSOR FUSION AND STATISTICAL ALGORITHMS Proceedings of the 2009 ASME International Mechanical Engineering Congress and Exposition IMECE2009 November 13-19, 2009, Lake Buena Vista, Florida, USA IMECE2009-12012 ENHANCED VEHICLE IDENTIFICATION

More information

Introduction to Deep Reinforcement Learning and Control

Introduction to Deep Reinforcement Learning and Control Carnegie Mellon School of Computer Science Deep Reinforcement Learning and Control Introduction to Deep Reinforcement Learning and Control Lecture 1, CMU 10703 Katerina Fragkiadaki Logistics 3 assignments

More information

Cognitive Modeling. Lecture 12: Bayesian Inference. Sharon Goldwater. School of Informatics University of Edinburgh

Cognitive Modeling. Lecture 12: Bayesian Inference. Sharon Goldwater. School of Informatics University of Edinburgh Cognitive Modeling Lecture 12: Bayesian Inference Sharon Goldwater School of Informatics University of Edinburgh sgwater@inf.ed.ac.uk February 18, 20 Sharon Goldwater Cognitive Modeling 1 1 Prediction

More information

Lecture 13: Finding optimal treatment policies

Lecture 13: Finding optimal treatment policies MACHINE LEARNING FOR HEALTHCARE 6.S897, HST.S53 Lecture 13: Finding optimal treatment policies Prof. David Sontag MIT EECS, CSAIL, IMES (Thanks to Peter Bodik for slides on reinforcement learning) Outline

More information

Georgetown University ECON-616, Fall Macroeconometrics. URL: Office Hours: by appointment

Georgetown University ECON-616, Fall Macroeconometrics.   URL:  Office Hours: by appointment Georgetown University ECON-616, Fall 2016 Macroeconometrics Instructor: Ed Herbst E-mail: ed.herbst@gmail.com URL: http://edherbst.net/ Office Hours: by appointment Scheduled Class Time and Organization:

More information

Bayesian integration in sensorimotor learning

Bayesian integration in sensorimotor learning Bayesian integration in sensorimotor learning Introduction Learning new motor skills Variability in sensors and task Tennis: Velocity of ball Not all are equally probable over time Increased uncertainty:

More information

Artificial Intelligence Lecture 7

Artificial Intelligence Lecture 7 Artificial Intelligence Lecture 7 Lecture plan AI in general (ch. 1) Search based AI (ch. 4) search, games, planning, optimization Agents (ch. 8) applied AI techniques in robots, software agents,... Knowledge

More information

Bayesian and Frequentist Approaches

Bayesian and Frequentist Approaches Bayesian and Frequentist Approaches G. Jogesh Babu Penn State University http://sites.stat.psu.edu/ babu http://astrostatistics.psu.edu All models are wrong But some are useful George E. P. Box (son-in-law

More information

EECS 433 Statistical Pattern Recognition

EECS 433 Statistical Pattern Recognition EECS 433 Statistical Pattern Recognition Ying Wu Electrical Engineering and Computer Science Northwestern University Evanston, IL 60208 http://www.eecs.northwestern.edu/~yingwu 1 / 19 Outline What is Pattern

More information

Computational Capacity and Statistical Inference: A Never Ending Interaction. Finbarr Sloane EHR/DRL

Computational Capacity and Statistical Inference: A Never Ending Interaction. Finbarr Sloane EHR/DRL Computational Capacity and Statistical Inference: A Never Ending Interaction Finbarr Sloane EHR/DRL Studies in Crop Variation I (1921) It has been estimated that Sir Ronald A. Fisher spent about 185

More information

Learning Navigational Maps by Observing the Movement of Crowds

Learning Navigational Maps by Observing the Movement of Crowds Learning Navigational Maps by Observing the Movement of Crowds Simon T. O Callaghan Australia, NSW s.ocallaghan@acfr.usyd.edu.au Surya P. N. Singh Australia, NSW spns@acfr.usyd.edu.au Fabio T. Ramos Australia,

More information

CS 4365: Artificial Intelligence Recap. Vibhav Gogate

CS 4365: Artificial Intelligence Recap. Vibhav Gogate CS 4365: Artificial Intelligence Recap Vibhav Gogate Exam Topics Search BFS, DFS, UCS, A* (tree and graph) Completeness and Optimality Heuristics: admissibility and consistency CSPs Constraint graphs,

More information

Lecture 9: The Agent Form of Bayesian Games

Lecture 9: The Agent Form of Bayesian Games Microeconomics I: Game Theory Lecture 9: The Agent Form of Bayesian Games (see Osborne, 2009, Sect 9.2.2) Dr. Michael Trost Department of Applied Microeconomics December 20, 2013 Dr. Michael Trost Microeconomics

More information

High-level Vision. Bernd Neumann Slides for the course in WS 2004/05. Faculty of Informatics Hamburg University Germany

High-level Vision. Bernd Neumann Slides for the course in WS 2004/05. Faculty of Informatics Hamburg University Germany High-level Vision Bernd Neumann Slides for the course in WS 2004/05 Faculty of Informatics Hamburg University Germany neumann@informatik.uni-hamburg.de http://kogs-www.informatik.uni-hamburg.de 1 Contents

More information

CS221 / Autumn 2017 / Liang & Ermon. Lecture 19: Conclusion

CS221 / Autumn 2017 / Liang & Ermon. Lecture 19: Conclusion CS221 / Autumn 2017 / Liang & Ermon Lecture 19: Conclusion Outlook AI is everywhere: IT, transportation, manifacturing, etc. AI being used to make decisions for: education, credit, employment, advertising,

More information

Lecture 3: Bayesian Networks 1

Lecture 3: Bayesian Networks 1 Lecture 3: Bayesian Networks 1 Jingpeng Li 1 Content Reminder from previous lecture: Bayes theorem Bayesian networks Why are they currently interesting? Detailed example from medical diagnostics Bayesian

More information

Dimensionality of the Force Concept Inventory: Comparing Bayesian Item Response Models. Xiaowen Liu Eric Loken University of Connecticut

Dimensionality of the Force Concept Inventory: Comparing Bayesian Item Response Models. Xiaowen Liu Eric Loken University of Connecticut Dimensionality of the Force Concept Inventory: Comparing Bayesian Item Response Models Xiaowen Liu Eric Loken University of Connecticut 1 Overview Force Concept Inventory Bayesian implementation of one-

More information

Continuous/Discrete Non Parametric Bayesian Belief Nets with UNICORN and UNINET

Continuous/Discrete Non Parametric Bayesian Belief Nets with UNICORN and UNINET Continuous/Discrete Non Parametric Bayesian Belief Nets with UNICORN and UNINET R.M. Cooke 1, D. Kurowicka, A. M. Hanea, O. Morales, D. A. Ababei Department of Mathematics Delft University of Technology

More information

Econometrics II - Time Series Analysis

Econometrics II - Time Series Analysis University of Pennsylvania Economics 706, Spring 2008 Econometrics II - Time Series Analysis Instructor: Frank Schorfheide; Room 525, McNeil Building E-mail: schorf@ssc.upenn.edu URL: http://www.econ.upenn.edu/

More information

PCA Enhanced Kalman Filter for ECG Denoising

PCA Enhanced Kalman Filter for ECG Denoising IOSR Journal of Electronics & Communication Engineering (IOSR-JECE) ISSN(e) : 2278-1684 ISSN(p) : 2320-334X, PP 06-13 www.iosrjournals.org PCA Enhanced Kalman Filter for ECG Denoising Febina Ikbal 1, Prof.M.Mathurakani

More information

Russian Journal of Agricultural and Socio-Economic Sciences, 3(15)

Russian Journal of Agricultural and Socio-Economic Sciences, 3(15) ON THE COMPARISON OF BAYESIAN INFORMATION CRITERION AND DRAPER S INFORMATION CRITERION IN SELECTION OF AN ASYMMETRIC PRICE RELATIONSHIP: BOOTSTRAP SIMULATION RESULTS Henry de-graft Acquah, Senior Lecturer

More information

Vision as Bayesian inference: analysis by synthesis?

Vision as Bayesian inference: analysis by synthesis? Vision as Bayesian inference: analysis by synthesis? Schwarz Andreas, Wiesner Thomas 1 / 70 Outline Introduction Motivation Problem Description Bayesian Formulation Generative Models Letters, Text Faces

More information

ERA: Architectures for Inference

ERA: Architectures for Inference ERA: Architectures for Inference Dan Hammerstrom Electrical And Computer Engineering 7/28/09 1 Intelligent Computing In spite of the transistor bounty of Moore s law, there is a large class of problems

More information

MITOCW conditional_probability

MITOCW conditional_probability MITOCW conditional_probability You've tested positive for a rare and deadly cancer that afflicts 1 out of 1000 people, based on a test that is 99% accurate. What are the chances that you actually have

More information

Partially-Observable Markov Decision Processes as Dynamical Causal Models. Finale Doshi-Velez NIPS Causality Workshop 2013

Partially-Observable Markov Decision Processes as Dynamical Causal Models. Finale Doshi-Velez NIPS Causality Workshop 2013 Partially-Observable Markov Decision Processes as Dynamical Causal Models Finale Doshi-Velez NIPS Causality Workshop 2013 The POMDP Mindset We poke the world (perform an action) Agent World The POMDP Mindset

More information

Model Based Probabilistic Inference for Intensive Care Medicine

Model Based Probabilistic Inference for Intensive Care Medicine 1 Model Based Probabilistic Inference for Intensive Care Medicine Yusuf B. Erol 1, Romi Phadte 1, Harsimran Sammy Sidhu 1, Claire Asselstine 1, David Phillips 2, Geoffrey Manley 2, and Stuart Russell 1

More information

Kelvin Chan Feb 10, 2015

Kelvin Chan Feb 10, 2015 Underestimation of Variance of Predicted Mean Health Utilities Derived from Multi- Attribute Utility Instruments: The Use of Multiple Imputation as a Potential Solution. Kelvin Chan Feb 10, 2015 Outline

More information

Individual Differences in Attention During Category Learning

Individual Differences in Attention During Category Learning Individual Differences in Attention During Category Learning Michael D. Lee (mdlee@uci.edu) Department of Cognitive Sciences, 35 Social Sciences Plaza A University of California, Irvine, CA 92697-5 USA

More information

Is Motion Planning Overrated? Jeannette Bohg - Interactive Perception and Robot Learning Lab - Stanford

Is Motion Planning Overrated? Jeannette Bohg - Interactive Perception and Robot Learning Lab - Stanford Is Motion Planning Overrated? Jeannette Bohg - Interactive Perception and Robot Learning Lab - Stanford Is Motion Planning Overrated? Jeannette Bohg - Interactive Perception and Robot Learning Lab - Stanford

More information

A Situated Model for Sensory-motor Coordination in Gaze Control

A Situated Model for Sensory-motor Coordination in Gaze Control A Situated Model for Sensory-motor Coordination in Gaze Control G. de Croon, E.O. Postma, H.J. van den Herik IKAT, Universiteit Maastricht, P.O. Box 616, 6200 MD, Maastricht, The Netherlands. Voice: 0031-433883477

More information

How do we combine two treatment arm trials with multiple arms trials in IPD metaanalysis? An Illustration with College Drinking Interventions

How do we combine two treatment arm trials with multiple arms trials in IPD metaanalysis? An Illustration with College Drinking Interventions 1/29 How do we combine two treatment arm trials with multiple arms trials in IPD metaanalysis? An Illustration with College Drinking Interventions David Huh, PhD 1, Eun-Young Mun, PhD 2, & David C. Atkins,

More information

The Classification Accuracy of Measurement Decision Theory. Lawrence Rudner University of Maryland

The Classification Accuracy of Measurement Decision Theory. Lawrence Rudner University of Maryland Paper presented at the annual meeting of the National Council on Measurement in Education, Chicago, April 23-25, 2003 The Classification Accuracy of Measurement Decision Theory Lawrence Rudner University

More information

Look-Ahead Monte Carlo with People

Look-Ahead Monte Carlo with People Look-Ahead Monte Carlo with People Charles Blundell (c.blundell@gatsby.ucl.ac.uk) Gatsby Computational Neuroscience Unit, University College London, London, UK Adam Sanborn (a.n.sanborn@warwick.ac.uk)

More information

ICS 606. Intelligent Autonomous Agents 1. Intelligent Autonomous Agents ICS 606 / EE 606 Fall Reactive Architectures

ICS 606. Intelligent Autonomous Agents 1. Intelligent Autonomous Agents ICS 606 / EE 606 Fall Reactive Architectures Intelligent Autonomous Agents ICS 606 / EE 606 Fall 2011 Nancy E. Reed nreed@hawaii.edu 1 Lecture #5 Reactive and Hybrid Agents Reactive Architectures Brooks and behaviors The subsumption architecture

More information

PHM for Rail Maintenance Management: at the crossroads between data science and physics

PHM for Rail Maintenance Management: at the crossroads between data science and physics PHM for Rail Maintenance Management: at the crossroads between data science and physics Pierre Dersin, Ph.D. PHM Director, ALSTOM Digital Mobility 15/09/2016 Presentation outline Introduction: the benefits

More information

IE 5203 Decision Analysis Lab I Probabilistic Modeling, Inference and Decision Making with Netica

IE 5203 Decision Analysis Lab I Probabilistic Modeling, Inference and Decision Making with Netica IE 5203 Decision Analysis Lab I Probabilistic Modeling, Inference and Decision Making with Netica Overview of Netica Software Netica Application is a comprehensive tool for working with Bayesian networks

More information

CISC453 Winter Probabilistic Reasoning Part B: AIMA3e Ch

CISC453 Winter Probabilistic Reasoning Part B: AIMA3e Ch CISC453 Winter 2010 Probabilistic Reasoning Part B: AIMA3e Ch 14.5-14.8 Overview 2 a roundup of approaches from AIMA3e 14.5-14.8 14.5 a survey of approximate methods alternatives to the direct computing

More information

Bayesian Analysis by Simulation

Bayesian Analysis by Simulation 408 Resampling: The New Statistics CHAPTER 25 Bayesian Analysis by Simulation Simple Decision Problems Fundamental Problems In Statistical Practice Problems Based On Normal And Other Distributions Conclusion

More information

5/20/2014. Leaving Andy Clark's safe shores: Scaling Predictive Processing to higher cognition. ANC sysmposium 19 May 2014

5/20/2014. Leaving Andy Clark's safe shores: Scaling Predictive Processing to higher cognition. ANC sysmposium 19 May 2014 ANC sysmposium 19 May 2014 Lorentz workshop on HPI Leaving Andy Clark's safe shores: Scaling Predictive Processing to higher cognition Johan Kwisthout, Maria Otworowska, Harold Bekkering, Iris van Rooij

More information

A formal analysis of cultural evolution by replacement

A formal analysis of cultural evolution by replacement A formal analysis of cultural evolution by replacement Jing Xu (jing.xu@berkeley.edu) Florencia Reali (floreali@berkeley.edu) Thomas L. Griffiths (tom griffiths@berkeley.edu) Department of Psychology,

More information

Lecture 4: Research Approaches

Lecture 4: Research Approaches Lecture 4: Research Approaches Lecture Objectives Theories in research Research design approaches ú Experimental vs. non-experimental ú Cross-sectional and longitudinal ú Descriptive approaches How to

More information

The random variable must be a numeric measure resulting from the outcome of a random experiment.

The random variable must be a numeric measure resulting from the outcome of a random experiment. Now we will define, discuss, and apply random variables. This will utilize and expand upon what we have already learned about probability and will be the foundation of the bridge between probability and

More information

Bayesian Networks Representation. Machine Learning 10701/15781 Carlos Guestrin Carnegie Mellon University

Bayesian Networks Representation. Machine Learning 10701/15781 Carlos Guestrin Carnegie Mellon University Bayesian Networks Representation Machine Learning 10701/15781 Carlos Guestrin Carnegie Mellon University March 16 th, 2005 Handwriting recognition Character recognition, e.g., kernel SVMs r r r a c r rr

More information

Introduction to Computational Neuroscience

Introduction to Computational Neuroscience Introduction to Computational Neuroscience Lecture 5: Data analysis II Lesson Title 1 Introduction 2 Structure and Function of the NS 3 Windows to the Brain 4 Data analysis 5 Data analysis II 6 Single

More information

MOBILE & SERVICE ROBOTICS RO OBOTIC CA 01. Supervision and control

MOBILE & SERVICE ROBOTICS RO OBOTIC CA 01. Supervision and control CY 02CFIC CFIDV MOBILE & SERVICE ROBOTICS Supervision and control Basilio Bona DAUIN Politecnico di Torino Basilio Bona DAUIN Politecnico di Torino 001/1 Supervision and Control 02CFIC CY a priori knowledge

More information

This is a repository copy of Feeling the Shape: Active Exploration Behaviors for Object Recognition With a Robotic Hand.

This is a repository copy of Feeling the Shape: Active Exploration Behaviors for Object Recognition With a Robotic Hand. This is a repository copy of Feeling the Shape: Active Exploration Behaviors for Object Recognition With a Robotic Hand. White Rose Research Online URL for this paper: http://eprints.whiterose.ac.uk/121310/

More information

BREAST CANCER EPIDEMIOLOGY MODEL:

BREAST CANCER EPIDEMIOLOGY MODEL: BREAST CANCER EPIDEMIOLOGY MODEL: Calibrating Simulations via Optimization Michael C. Ferris, Geng Deng, Dennis G. Fryback, Vipat Kuruchittham University of Wisconsin 1 University of Wisconsin Breast Cancer

More information

Slide 1 - Introduction to Statistics Tutorial: An Overview Slide notes

Slide 1 - Introduction to Statistics Tutorial: An Overview Slide notes Slide 1 - Introduction to Statistics Tutorial: An Overview Introduction to Statistics Tutorial: An Overview. This tutorial is the first in a series of several tutorials that introduce probability and statistics.

More information

Rethinking Cognitive Architecture!

Rethinking Cognitive Architecture! Rethinking Cognitive Architecture! Reconciling Uniformity and Diversity via Graphical Models! Paul Rosenbloom!!! 1/25/2010! Department of Computer Science &! Institute for Creative Technologies! The projects

More information

Bayesian Perception & Decision for Intelligent Mobility

Bayesian Perception & Decision for Intelligent Mobility Bayesian Perception & Decision for Intelligent Mobility E-Motion & Chroma teams Inria Research Center Grenoble Rhône-Alpes Christian LAUGIER First Class Research Director at Inria San Francisco, 05/11/2015

More information

Supplementary notes for lecture 8: Computational modeling of cognitive development

Supplementary notes for lecture 8: Computational modeling of cognitive development Supplementary notes for lecture 8: Computational modeling of cognitive development Slide 1 Why computational modeling is important for studying cognitive development. Let s think about how to study the

More information

BAYESIAN NETWORK FOR FAULT DIAGNOSIS

BAYESIAN NETWORK FOR FAULT DIAGNOSIS BAYESIAN NETWOK FO FAULT DIAGNOSIS C.H. Lo, Y.K. Wong and A.B. ad Department of Electrical Engineering, The Hong Kong Polytechnic University Hung Hom, Kowloon, Hong Kong Fax: +852 2330 544 Email: eechlo@inet.polyu.edu.hk,

More information

Appendix I Teaching outcomes of the degree programme (art. 1.3)

Appendix I Teaching outcomes of the degree programme (art. 1.3) Appendix I Teaching outcomes of the degree programme (art. 1.3) The Master graduate in Computing Science is fully acquainted with the basic terms and techniques used in Computing Science, and is familiar

More information

POND-Hindsight: Applying Hindsight Optimization to POMDPs

POND-Hindsight: Applying Hindsight Optimization to POMDPs POND-Hindsight: Applying Hindsight Optimization to POMDPs Alan Olsen and Daniel Bryce alan@olsen.org, daniel.bryce@usu.edu Utah State University Logan, UT Abstract We present the POND-Hindsight entry in

More information

An Ideal Observer Model of Visual Short-Term Memory Predicts Human Capacity Precision Tradeoffs

An Ideal Observer Model of Visual Short-Term Memory Predicts Human Capacity Precision Tradeoffs An Ideal Observer Model of Visual Short-Term Memory Predicts Human Capacity Precision Tradeoffs Chris R. Sims Robert A. Jacobs David C. Knill (csims@cvs.rochester.edu) (robbie@bcs.rochester.edu) (knill@cvs.rochester.edu)

More information

Approximate Inference in Bayes Nets Sampling based methods. Mausam (Based on slides by Jack Breese and Daphne Koller)

Approximate Inference in Bayes Nets Sampling based methods. Mausam (Based on slides by Jack Breese and Daphne Koller) Approximate Inference in Bayes Nets Sampling based methods Mausam (Based on slides by Jack Breese and Daphne Koller) 1 Bayes Nets is a generative model We can easily generate samples from the distribution

More information

Behavioral Data Mining. Lecture 4 Measurement

Behavioral Data Mining. Lecture 4 Measurement Behavioral Data Mining Lecture 4 Measurement Outline Hypothesis testing Parametric statistical tests Non-parametric tests Precision-Recall plots ROC plots Hardware update Icluster machines are ready for

More information

PSYCH-GA.2211/NEURL-GA.2201 Fall 2016 Mathematical Tools for Cognitive and Neural Science. Homework 5

PSYCH-GA.2211/NEURL-GA.2201 Fall 2016 Mathematical Tools for Cognitive and Neural Science. Homework 5 PSYCH-GA.2211/NEURL-GA.2201 Fall 2016 Mathematical Tools for Cognitive and Neural Science Homework 5 Due: 21 Dec 2016 (late homeworks penalized 10% per day) See the course web site for submission details.

More information

A Bayesian Nonparametric Model Fit statistic of Item Response Models

A Bayesian Nonparametric Model Fit statistic of Item Response Models A Bayesian Nonparametric Model Fit statistic of Item Response Models Purpose As more and more states move to use the computer adaptive test for their assessments, item response theory (IRT) has been widely

More information

Geoffrey Stewart Morrison

Geoffrey Stewart Morrison Basic Workshop on Forensic Statistics Introduction to logical reasoning for the evaluation of forensic evidence Geoffrey Stewart Morrison p(e H ) p p(e H ) d These slides are available on the following

More information

Intelligent Agents. Soleymani. Artificial Intelligence: A Modern Approach, Chapter 2

Intelligent Agents. Soleymani. Artificial Intelligence: A Modern Approach, Chapter 2 Intelligent Agents CE417: Introduction to Artificial Intelligence Sharif University of Technology Spring 2016 Soleymani Artificial Intelligence: A Modern Approach, Chapter 2 Outline Agents and environments

More information

Burn-in, bias, and the rationality of anchoring

Burn-in, bias, and the rationality of anchoring Burn-in, bias, and the rationality of anchoring Falk Lieder Translational Neuromodeling Unit ETH and University of Zurich lieder@biomed.ee.ethz.ch Thomas L. Griffiths Psychology Department University of

More information

A Decision-Theoretic Approach to Evaluating Posterior Probabilities of Mental Models

A Decision-Theoretic Approach to Evaluating Posterior Probabilities of Mental Models A Decision-Theoretic Approach to Evaluating Posterior Probabilities of Mental Models Jonathan Y. Ito and David V. Pynadath and Stacy C. Marsella Information Sciences Institute, University of Southern California

More information

A Hierarchical Adaptive Approach to the Optimal Design of Experiments

A Hierarchical Adaptive Approach to the Optimal Design of Experiments A Hierarchical Adaptive Approach to the Optimal Design of Experiments Woojae Kim 1 (kim.1124@osu.edu), Mark Pitt 1 (pitt.2@osu.edu), Zhong-Lin Lu 1 (lu.535@osu.edu), Mark Steyvers 2 (mark.steyvers@uci.edu),

More information

Bayesian Confidence Intervals for Means and Variances of Lognormal and Bivariate Lognormal Distributions

Bayesian Confidence Intervals for Means and Variances of Lognormal and Bivariate Lognormal Distributions Bayesian Confidence Intervals for Means and Variances of Lognormal and Bivariate Lognormal Distributions J. Harvey a,b, & A.J. van der Merwe b a Centre for Statistical Consultation Department of Statistics

More information

NeuroSky s esense Meters and Detection of Mental State

NeuroSky s esense Meters and Detection of Mental State NeuroSky s esense Meters and Detection of Mental State The Attention and Meditation esense meters output by NeuroSky s MindSet are comprised of a complex combination of artifact rejection and data classification

More information

Bayesian (Belief) Network Models,

Bayesian (Belief) Network Models, Bayesian (Belief) Network Models, 2/10/03 & 2/12/03 Outline of This Lecture 1. Overview of the model 2. Bayes Probability and Rules of Inference Conditional Probabilities Priors and posteriors Joint distributions

More information

Learning to Identify Irrelevant State Variables

Learning to Identify Irrelevant State Variables Learning to Identify Irrelevant State Variables Nicholas K. Jong Department of Computer Sciences University of Texas at Austin Austin, Texas 78712 nkj@cs.utexas.edu Peter Stone Department of Computer Sciences

More information

Multi-agent Probabilistic Search in a Sequential Decision-theoretic Framework

Multi-agent Probabilistic Search in a Sequential Decision-theoretic Framework 2008 IEEE International Conference on Robotics and Automation Pasadena, CA, USA, May 19-23, 2008 Multi-agent Probabilistic Search in a Sequential Decision-theoretic Framework Timothy H. Chung Department

More information

FrD /07/$ IEEE. 4386

FrD /07/$ IEEE. 4386 2007 IEEE International Conference on Robotics and Automation Roma, Italy, 10-14 April 2007 FrD5.3 A Decision-Making Framework for Control Strategies in Probabilistic Search Timothy H. Chung and Joel W.

More information

Running head: RATIONAL APPROXIMATIONS TO CATEGORY LEARNING. Rational approximations to rational models: Alternative algorithms for category learning

Running head: RATIONAL APPROXIMATIONS TO CATEGORY LEARNING. Rational approximations to rational models: Alternative algorithms for category learning Rational Approximations to Category Learning 1 Running head: RATIONAL APPROXIMATIONS TO CATEGORY LEARNING Rational approximations to rational models: Alternative algorithms for category learning Adam N.

More information

MTH 225: Introductory Statistics

MTH 225: Introductory Statistics Marshall University College of Science Mathematics Department MTH 225: Introductory Statistics Course catalog description Basic probability, descriptive statistics, fundamental statistical inference procedures

More information

Lecture 2 Agents & Environments (Chap. 2) Outline

Lecture 2 Agents & Environments (Chap. 2) Outline Lecture 2 Agents & Environments (Chap. 2) Based on slides by UW CSE AI faculty, Dan Klein, Stuart Russell, Andrew Moore Outline Agents and environments Rationality PEAS specification Environment types

More information

A simple introduction to Markov Chain Monte Carlo sampling

A simple introduction to Markov Chain Monte Carlo sampling DOI 10.3758/s13423-016-1015-8 BRIEF REPORT A simple introduction to Markov Chain Monte Carlo sampling Don van Ravenzwaaij 1,2 Pete Cassey 2 Scott D. Brown 2 The Author(s) 2016. This article is published

More information

Persistent Visual Tracking and Accurate Geo-Location of Moving Ground Targets by Small Air Vehicles

Persistent Visual Tracking and Accurate Geo-Location of Moving Ground Targets by Small Air Vehicles Persistent Visual Tracking and Accurate Geo-Location of Moving Ground Targets by Small Air Vehicles Michael Dille, Ben Grocholsky, Stephen Thomas Nuske EXTENDED ABSTRACT 1 Motivation, Problem Statement,

More information

Michael Hallquist, Thomas M. Olino, Paul A. Pilkonis University of Pittsburgh

Michael Hallquist, Thomas M. Olino, Paul A. Pilkonis University of Pittsburgh Comparing the evidence for categorical versus dimensional representations of psychiatric disorders in the presence of noisy observations: a Monte Carlo study of the Bayesian Information Criterion and Akaike

More information

Bayes theorem describes the probability of an event, based on prior knowledge of conditions that might be related to the event.

Bayes theorem describes the probability of an event, based on prior knowledge of conditions that might be related to the event. Bayes theorem Bayes' Theorem is a theorem of probability theory originally stated by the Reverend Thomas Bayes. It can be seen as a way of understanding how the probability that a theory is true is affected

More information

CS 771 Artificial Intelligence. Intelligent Agents

CS 771 Artificial Intelligence. Intelligent Agents CS 771 Artificial Intelligence Intelligent Agents What is AI? Views of AI fall into four categories 1. Thinking humanly 2. Acting humanly 3. Thinking rationally 4. Acting rationally Acting/Thinking Humanly/Rationally

More information

INTRODUCTION TO PSYCHOLOGY

INTRODUCTION TO PSYCHOLOGY INTRODUCTION TO PSYCHOLOGY SUMMARY 1 ABDULLAH ALZIBDEH Introduction In this lecture, we discuss the definitions of psychology and behavior. We also discuss the approaches in psychology and the scientific

More information

You must answer question 1.

You must answer question 1. Research Methods and Statistics Specialty Area Exam October 28, 2015 Part I: Statistics Committee: Richard Williams (Chair), Elizabeth McClintock, Sarah Mustillo You must answer question 1. 1. Suppose

More information

An Application of Inverse Reinforcement Learning to Medical Records of Diabetes Treatment

An Application of Inverse Reinforcement Learning to Medical Records of Diabetes Treatment An Application of Inverse Reinforcement Learning to Medical Records of Diabetes Treatment H. Asoh, M. Shiro, S. Akaho, T. Kamishima (National Institute of Advanced Industrial Science and Technology) K.

More information

SLAUGHTER PIG MARKETING MANAGEMENT: UTILIZATION OF HIGHLY BIASED HERD SPECIFIC DATA. Henrik Kure

SLAUGHTER PIG MARKETING MANAGEMENT: UTILIZATION OF HIGHLY BIASED HERD SPECIFIC DATA. Henrik Kure SLAUGHTER PIG MARKETING MANAGEMENT: UTILIZATION OF HIGHLY BIASED HERD SPECIFIC DATA Henrik Kure Dina, The Royal Veterinary and Agricuural University Bülowsvej 48 DK 1870 Frederiksberg C. kure@dina.kvl.dk

More information

CS148 - Building Intelligent Robots Lecture 5: Autonomus Control Architectures. Instructor: Chad Jenkins (cjenkins)

CS148 - Building Intelligent Robots Lecture 5: Autonomus Control Architectures. Instructor: Chad Jenkins (cjenkins) Lecture 5 Control Architectures Slide 1 CS148 - Building Intelligent Robots Lecture 5: Autonomus Control Architectures Instructor: Chad Jenkins (cjenkins) Lecture 5 Control Architectures Slide 2 Administrivia

More information