A stochastic model-based approach to online event prediction and response scheduling

Size: px
Start display at page:

Download "A stochastic model-based approach to online event prediction and response scheduling"

Transcription

1 A stochastic model-based approach to online event prediction and response scheduling M. Biagi, L. Carnevali, M. Paolieri, F. Patara, E. Vicario Department of Information Engineering, University of Florence, Italy Chios, EPEW October 5, 26 This is about: prediction and scheduling in partially observable systems with a stochastic model-based approach / 23

2 Aim Partially observable system: any system where part of the state remains hidden to observations Diagnosis: estimate current state based on observations Prediction: evaluate future state probabilities given a distribution of initial current state Scheduling: select an action type and a time point in the future 2 / 23

3 An application context Ambient Assisted Living (AAL): aims at providing assistance to people living within smart environments through the integration and exploitation of new sensing technologies, data processing techniques, and services 3 / 23

4 An application context T.vanKasteren, A.Noulas, G.Englebienne, B.Krse, Accurate Activity Recognition in a Home Setting, UbiComp 8 4 / 23

5 Diagnosis given a stream of Observations (EventType and Timestamp), evaluate the current ongoing Activity of the subject and a measure of likelihood of such Activity diagnosis 2 provides a list of triplets π, x, R(t) where π [, ] is a likelihood measure x S is a state of the system R(t) is the PDF of the remaining sojourn time in x (estimated numerically on a grid of time points). 2 L.Carnevali, C.Nugent, F.Patara, E.Vicario, A Continuous-Time Model-Based Approach to Activity Recognition for Ambient Assisted Living QEST 5 5 / 23

6 Prediction an activity type is chosen as target the aim is to predict when the next instance of the target activity type will begin at each new diagnosis result, estimate the first-passage transient probabilities to the target activity 6 / 23

7 Prediction technique: a model for prediction model-based approach a predefined parametric structure: stochastic parameters derived automatically from the statistics in process mining terminology: enhancement without elicitation p xy idle xy act x idle yx p yx act y p xz idle zx idle p zy p yz zx p zy idle xz act z idle yz 7 / 23

8 Prediction technique: a model for prediction model structure is chosen so as to fit prediction needs take into account GEN state sojourn time distribution idle states are specialized, this allows us to: preserve memory of the foregoing activity (trade-off memory/complexity) duration depends on previous and following activities. Allows to have idles durations with a lower CV that is a good feature for prediction never observed states (e.g from dinner to dinner) are not included allowing to reduce the space of possible behaviors p xy idle xy act x idle yx p yx act y p xz idle zx idle p zy p yz zx p zy idle xz act z idle yz 8 / 23

9 Prediction technique: derivation of stochastic parameters sojourn time distributions are evaluated via a basic approximants fitting of expected value and variation coefficient 3 for CV >, a mixture of 2 EXP for > CV > 2 a sequence of 2 EXP for 2 CV, a shifted EXP transition probabilities p xy that activity act y will follow act x is estimated as p xy = { i < n α i = act x and α i+ = act y } { i < n α i = act x } 3 W. Whitt, Approximating a Point Process by a Renewal Process, I: Two Basic Methods, Operations Research, / 23

10 Prediction technique: distribution to critical event π A =.7 A D E π B =.3 B C F for each current plausible state (A and B) evaluate the first-passage distribution to the critical state (gray state) for each possible path to the critical state, compose the remaining time in the current activity with the Semi Markov sojourn in subsequent activities / 23

11 Prediction technique: distribution to critical event π A =.7 A D E π B =.3 B C F for each current plausible state (A and B) evaluate the first-passage distribution to the critical state (gray state) 2 sum up first passage time distributions weighted by the diagnosed state likelihood PDF (t) A to F π A PDF (t) B to F π B PDF (t) to F = t t t / 23

12 Scheduling Select and schedule a response action to be taken before or at the beginning of some target activity (e.g.: remind pills intake 5 minutes before a meal) 2 / 23

13 Scheduling: problem formulation 4 Response actuation time: response action must be executed t w time units before the beginning of the target activity. Response duration: response action is active for a fixed duration : target activities are successfully handled by the system if they start t w time units after the response action was active. dinner t ev t t w time E.g. pills intake reminder: t w = 3s = 2s Remind at time t 4 F. Salfner et al., A Survey of Online Failure Prediction Methods, ACM Comput. Surv., 2 3 / 23

14 Scheduling: reward and decision policy performance evaluation: based on the number of true positives TP, false positives FP and false negatives FN maximum probability interval policy: activation time is evaluated searching for the maximum probability interval of width if maximum probability is higher than a fixed threshold, schedule at t = x t w, where x is the start of the interval otherwise, no response action is scheduled dinner t ev t PDF (t) x t w t time 4 / 23

15 Scheduling: an example a = α, τ, δ e τ t τ + δ t () PDF (t) t w t a 2 = α 2, τ 2, δ 2 τ 2 τ 2 + δ 2 time 5 / 23

16 Scheduling: an example e e 2 τ t t2 τ + δ PDF (t) a = α, τ, δ PDF 2 (t) t w e 3 t3 t (2) t a 2 = α 2, τ 2, δ 2 e 4 e 5 τ 2 t4 t5 τ 2 + δ 2 time t () Schedule t () is updated at e 2 t w PDF 3 (t) I t (3) t w I 2 t Schedule t (2) results in a FP Schedule t (3) results in a TP t 6 / 23

17 Experiments: methodology two types of datasets considered: synthetic datasets real AAL dataset 5 compare also different fitting strategies for sojourn distributions: the mean value with an EXP distribution the mean value with a two phases Erlang distribution the mean value and CV by the Whitt approach 6 exact state diagnosis assumed as input, allowing to focus on predictor and scheduler performance results as precision = TP/(TP + FP) and recall = TP/(TP + FN) 5 T.vanKasteren, A.Noulas, G.Englebienne, B.Krse, Accurate Activity Recognition in a Home Setting, UbiComp 8 6 W. Whitt, Approximating a Point Process by a Renewal Process, I: Two Basic Methods, Operations Research, / 23

18 Experiments: generating a synthetic datasets generated by simulation in order to analyze how real phenomena sojourn distributions affects system predictability predefined model structure x, y p x,y =.5 events with Exp() inter-times p, p,2 p 2,3 act idle, act idle,2 act 2 idle 2,3 act 3 p, p, p 2, p 3, idle, idle, idle 2, idle 3, 8-phase Erlang: Erlang(k = 8, λ = 4) µ = 2, CV = 8 2-phase Erlang: Erlang(k = 2, λ = ) µ = 2, CV = 2 2-phase Hyper-exp: hyper-exp µ = 2, CV.92 increase CV 8 / 23

19 Experiments: synthetic datasets results Precision t w = t w = 6 t w = 2 t w = Recall t w = t w = 6 t w = 2 t w = 3 precision and recall with Whitt fitting on 8-phase Erlang 9 / 23

20 Experiments: synthetic datasets results Whitt Erlang Exp t w = t w = 6 t w = 2 t w = t w = t w = 6 t w = 2 t w = t w = t w = 6 t w = 2 t w = comparing precisions of fitting strategy on 8-phase Erlang synthetic dataset 2 / 23

21 Experiments: synthetic datasets results ph Erlang CV = 8 2-ph Erlang CV = 2 Hyper-exp CV.92 t w = t w = 6 t w = 2 t w = t w = t w = 6 t w = 2 t w = t w = t w = 6 t w = 2 t w = comparing precisions with Whitt fitting on different synthetic sojourn distributions 2 / 23

22 Experiments: real AAL dataset results increase F F = 2 (precision recall) (precision+recall) Preparing dinner Preparing breakfast Preparing breakfast (filtered) (s) /.63/ / /.684 /.86 /.86 6 /.5/.25.2/.385.2/ /.78.68/.78 t w (s) 2 /.2/.385.2/.385.2/ /.78.68/.78 3 / / / /.48/.6.48/.6 t w (s) t w (s) /.93/ / / / / /.24/.36.24/.36.24/ / /.25 2 / / / /.24/.36.24/.36 3 / / / /.2/.36.2/.36 /.465/ / /.88.66/ /.93 6 /.222/ / / / / /.63/.26.28/ / / / /.53/.28.88/ / /.8.329/ / 23

23 Conclusion and future development we presented a model-based approach to online event prediction and response scheduling fitting both sojourn times mean value and CV improves precision of the prediction performance of the technique depends on the predictability of the system this technique can be applied to AAL for small values of response actuation time future evolution often depends on the history of previous activities, breaking the hypothesis of semi-markov models performance achieved by different model structures will be investigated in the future applicability beyond the limits of AAL perhaps better suited to more structured and predictable systems e.g. monitoring of industrial system e.g. gas/water distribution network 23 / 23

24 24 / 23

25 An application context Observed: Open fridge, open bathroom door Hidden states: Dinner, toileting,... Floorplan of the house, red rectangle boxes indicate sensor nodes 7 7 T.vanKasteren, A.Noulas, G.Englebienne, B.Krse, Accurate Activity Recognition in a Home Setting, UbiComp 8 25 / 23

26 A reference dataset for AAL a sequence of Observations with manually annotated Activities 8 7+ Activity Types from the index of Activities of Daily Living 28 Event Types (On Off of 4 state-change Sensors) 28 days, 245 annotated Activities and 29 unclassified intervals (Idle), 2638 Observations,... a Idling k- a Idling k a k+ δ δ k-,k k e n-u e n-u+ e n- e n e n+ e n+h end t k- t k start t k end start t k+ single subject, no concurrent Activities, reliable sensors, some inaccuracies start and completion of Activities do not coincide with Observations (and they are thus not observable) 8 T.vanKasteren, A.Noulas, G.Englebienne, B.Krse, Accurate Activity Recognition in a Home Setting, UbiComp 8 26 / 23

27 Prediction technique: a model for prediction Activity µ CV bed toilet breakfast shower leave drink dinner idle Activity µ CV idle bed idle toilet idle breakfast idle shower idle leave idle drink idle dinner Activity µ CV idle bed toilet idle bed breakfast idle toilet bed idle toilet toilet idle toilet breakfast idle toilet shower idle toilet leave idle toilerink idle toileinner idle breakfast toilet idle breakfast shower idle breakfasrink idle shower toilet 2.. idle shower leave idle shower dinner idle leave bed 22.. idle leave toilet idle leave shower idle leave leave idle leave drink idle leave dinner 76.. idle drink toilet idle drink breakfast 8.. idle drink drink idle dinner toilet idle dinner leave idle dinner drink / 23

28 Prediction technique: distribution to critical event π A =.7 A p AD D E p AC π B =.3 B C F for each current plausible state (A and B) evaluate the first-passage distribution to the critical state (gray state) 9 A. Horváth, M. Paolieri, L. Ridi, E.Vicario, Transient analysis of non-markovian models using stochastic state classes Perform. Evaluation, / 23

29 Prediction technique: distribution to critical event π A =.7 A p AD D E p AC π B =.3 B C F for each current plausible state (A and B) evaluate the first-passage distribution to the critical state (gray state) mark critical state as absorbing 9 A. Horváth, M. Paolieri, L. Ridi, E.Vicario, Transient analysis of non-markovian models using stochastic state classes Perform. Evaluation, / 23

30 Prediction technique: distribution to critical event π A =.7 A p AD D E p AC π B =.3 B C F for each current plausible state (A and B) evaluate the first-passage distribution to the critical state (gray state) mark critical state as absorbing 2 evaluate next states reached after current state (C and D from A) 9 A. Horváth, M. Paolieri, L. Ridi, E.Vicario, Transient analysis of non-markovian models using stochastic state classes Perform. Evaluation, / 23

31 Prediction technique: distribution to critical event π A =.7 A p AD D E p AC π B =.3 B C F for each current plausible state (A and B) evaluate the first-passage distribution to the critical state (gray state) mark critical state as absorbing 2 evaluate next states reached after current state (C and D from A) 3 P DF (t) and P CF (t) are the distributions from next states to the critical state and was evaluated offline with transient analysis 9 9 A. Horváth, M. Paolieri, L. Ridi, E.Vicario, Transient analysis of non-markovian models using stochastic state classes Perform. Evaluation, / 23

32 Prediction technique: distribution to critical event π A =.7 A p AD D E p AC π B =.3 B C F for each current plausible state (A and B) evaluate the first-passage distribution to the critical state (gray state) mark critical state as absorbing 2 evaluate next states reached after current state (C and D from A) 3 P DF (t) and P CF (t) are the distributions from next states to the critical state and was evaluated offline with transient analysis 9 4 evaluate convolution between R A (t) and P CF (t) weighted by p AC 9 A. Horváth, M. Paolieri, L. Ridi, E.Vicario, Transient analysis of non-markovian models using stochastic state classes Perform. Evaluation, / 23

33 Prediction technique: distribution to critical event π A =.7 A p AD D E p AC π B =.3 B C F for each current plausible state (A and B) evaluate the first-passage distribution to the critical state (gray state) mark critical state as absorbing 2 evaluate next states reached after current state (C and D from A) 3 P DF (t) and P CF (t) are the distributions from next states to the critical state and was evaluated offline with transient analysis 9 4 evaluate convolution between R A (t) and P CF (t) weighted by p AC 5 evaluate convolution between R A (t) and P DF (t) weighted by p AD 9 A. Horváth, M. Paolieri, L. Ridi, E.Vicario, Transient analysis of non-markovian models using stochastic state classes Perform. Evaluation, / 23

34 Prediction technique: distribution to critical event π A =.7 A p AD D E p AC π B =.3 B C F for each current plausible state (A and B) evaluate the first-passage distribution to the critical state (gray state) mark critical state as absorbing 2 evaluate next states reached after current state (C and D from A) 3 P DF (t) and P CF (t) are the distributions from next states to the critical state and was evaluated offline with transient analysis 9 4 evaluate convolution between R A (t) and P CF (t) weighted by p AC 5 evaluate convolution between R A (t) and P DF (t) weighted by p AD 6 their sum is the first passage time distribution from A to F 9 A. Horváth, M. Paolieri, L. Ridi, E.Vicario, Transient analysis of non-markovian models using stochastic state classes Perform. Evaluation, / 23

35 Experiments: synthetic datasets results Precision t w = t w = 6 t w = 2 t w = Recall t w = t w = 6 t w = 2 t w = Whitt t w = t w = 6 t w = 2 t w = t w = t w = 6 t w = 2 t w = EXP comparing fitting strategy on 8-phase Erlang synthetic dataset t w = t w = 6 t w = 2 t w = t w = t w = 6 t w = 2 t w = Erlang 35 / 23

36 Experiments: synthetic datasets results Precision t w = t w = 6 t w = 2 t w = Recall t w = t w = 6 t w = 2 t w = ph Erlang t w = t w = 6 t w = 2 t w = t w = t w = 6 t w = 2 t w = ph Erlang comparing results with Whitt fitting on different underlying distributions t t w = t w = 6 t w = 2 t w = t t w = t w = 6 t w = 2 t w = 3 Hyper-exp 36 / 23

A stochastic model-based approach to online event prediction and response scheduling

A stochastic model-based approach to online event prediction and response scheduling A stochastic model-based approach to online event prediction and response scheduling Marco Biagi, Laura Carnevali, Marco Paolieri, Fulvio Patara, and Enrico Vicario Department of Information Engineering,

More information

A framework for simulation and symbolic state space analysis of non-markovian models

A framework for simulation and symbolic state space analysis of non-markovian models A framework for simulation and symbolic state space analysis of non-markovian models Laura Carnevali, Lorenzo Ridi, Enrico Vicario SW Technologies Lab (STLab) - Dip. Sistemi e Informatica (DSI) - Univ.

More information

Probabilistic Deadline Miss Analysis of Real-Time Systems Using Regenerative Transient Analysis

Probabilistic Deadline Miss Analysis of Real-Time Systems Using Regenerative Transient Analysis Probabilistic Deadline Miss Analysis of Real-Time Systems Using Regenerative Transient Analysis L. Carnevali 1, A. Melani 2, L. Santinelli 3, G. Lipari 4 1 Department of Information Engineering, University

More information

An inspection-based compositional approach to the quantitative evaluation of assembly lines

An inspection-based compositional approach to the quantitative evaluation of assembly lines An inspection-based compositional approach to the quantitative evaluation of assembly lines Marco Biagi 1 Laura Carnevali 1 Tommaso Papini 1 Kumiko Tadano 2 Enrico Vicario 1 1 Department of Information

More information

Exploiting non-deterministic analysis in the integration of transient solution techniques for Markov Regenerative Processes

Exploiting non-deterministic analysis in the integration of transient solution techniques for Markov Regenerative Processes Exploiting non-deterministic analysis in the integration of transient solution techniques for Markov Regenerative Processes Marco Biagi 1, Laura Carnevali 1, Marco Paolieri, 2 Tommaso Papini 1, and Enrico

More information

Tennis player segmentation for semantic behavior analysis

Tennis player segmentation for semantic behavior analysis Proposta di Tennis player segmentation for semantic behavior analysis Architettura Software per Robot Mobili Vito Renò, Nicola Mosca, Massimiliano Nitti, Tiziana D Orazio, Donato Campagnoli, Andrea Prati,

More information

Stochastic Renewal Processes in Structural Reliability Analysis:

Stochastic Renewal Processes in Structural Reliability Analysis: Stochastic Renewal Processes in Structural Reliability Analysis: An Overview of Models and Applications Professor and Industrial Research Chair Department of Civil and Environmental Engineering University

More information

Lecturer: Olga Galinina

Lecturer: Olga Galinina Renewal models Lecturer: Olga Galinina E-mail: olga.galinina@tut.fi Outline Reminder. Exponential models definition of renewal processes exponential interval distribution Erlang distribution hyperexponential

More information

f X (x) = λe λx, , x 0, k 0, λ > 0 Γ (k) f X (u)f X (z u)du

f X (x) = λe λx, , x 0, k 0, λ > 0 Γ (k) f X (u)f X (z u)du 11 COLLECTED PROBLEMS Do the following problems for coursework 1. Problems 11.4 and 11.5 constitute one exercise leading you through the basic ruin arguments. 2. Problems 11.1 through to 11.13 but excluding

More information

Anomaly Detection for the CERN Large Hadron Collider injection magnets

Anomaly Detection for the CERN Large Hadron Collider injection magnets Anomaly Detection for the CERN Large Hadron Collider injection magnets Armin Halilovic KU Leuven - Department of Computer Science In cooperation with CERN 2018-07-27 0 Outline 1 Context 2 Data 3 Preprocessing

More information

Smart Home Health Analytics Information Systems University of Maryland Baltimore County

Smart Home Health Analytics Information Systems University of Maryland Baltimore County Smart Home Health Analytics Information Systems University of Maryland Baltimore County 1 IEEE Expert, October 1996 2 Given sample S from all possible examples D Learner L learns hypothesis h based on

More information

Lecture 4 Discriminant Analysis, k-nearest Neighbors

Lecture 4 Discriminant Analysis, k-nearest Neighbors Lecture 4 Discriminant Analysis, k-nearest Neighbors Fredrik Lindsten Division of Systems and Control Department of Information Technology Uppsala University. Email: fredrik.lindsten@it.uu.se fredrik.lindsten@it.uu.se

More information

ring structure Abstract Optical Grid networks allow many computing sites to share their resources by connecting

ring structure Abstract Optical Grid networks allow many computing sites to share their resources by connecting Markovian approximations for a grid computing network with a ring structure J. F. Pérez and B. Van Houdt Performance Analysis of Telecommunication Systems Research Group, Department of Mathematics and

More information

Markov localization uses an explicit, discrete representation for the probability of all position in the state space.

Markov localization uses an explicit, discrete representation for the probability of all position in the state space. Markov Kalman Filter Localization Markov localization localization starting from any unknown position recovers from ambiguous situation. However, to update the probability of all positions within the whole

More information

Mining State Dependencies Between Multiple Sensor Data Sources

Mining State Dependencies Between Multiple Sensor Data Sources Mining State Dependencies Between Multiple Sensor Data Sources C. Robardet Co-Authored with Marc Plantevit and Vasile-Marian Scuturici April 2013 1 / 27 Mining Sensor data A timely challenge? Why is it

More information

DeepLog: Anomaly Detection and Diagnosis from System Logs through Deep Learning. Min Du, Feifei Li, Guineng Zheng, Vivek Srikumar University of Utah

DeepLog: Anomaly Detection and Diagnosis from System Logs through Deep Learning. Min Du, Feifei Li, Guineng Zheng, Vivek Srikumar University of Utah DeepLog: Anomaly Detection and Diagnosis from System Logs through Deep Learning Min Du, Feifei Li, Guineng Zheng, Vivek Srikumar University of Utah Background 2 Background System Event Log 3 Background

More information

Environment (E) IBP IBP IBP 2 N 2 N. server. System (S) Adapter (A) ACV

Environment (E) IBP IBP IBP 2 N 2 N. server. System (S) Adapter (A) ACV The Adaptive Cross Validation Method - applied to polling schemes Anders Svensson and Johan M Karlsson Department of Communication Systems Lund Institute of Technology P. O. Box 118, 22100 Lund, Sweden

More information

Complexity of stochastic branch and bound methods for belief tree search in Bayesian reinforcement learning

Complexity of stochastic branch and bound methods for belief tree search in Bayesian reinforcement learning Complexity of stochastic branch and bound methods for belief tree search in Bayesian reinforcement learning Christos Dimitrakakis Informatics Institute, University of Amsterdam, Amsterdam, The Netherlands

More information

Abstract. 1 Introduction. Cointerpretation of Flow Rate-Pressure-Temperature Data from Permanent Downhole Gauges. Deconvolution. Breakpoint detection

Abstract. 1 Introduction. Cointerpretation of Flow Rate-Pressure-Temperature Data from Permanent Downhole Gauges. Deconvolution. Breakpoint detection Cointerpretation of Flow Rate-Pressure-Temperature Data from Permanent Downhole Gauges CS 229 Course Final Report Chuan Tian chuant@stanford.edu Yue Li yuel@stanford.edu Abstract This report documents

More information

CSL302/612 Artificial Intelligence End-Semester Exam 120 Minutes

CSL302/612 Artificial Intelligence End-Semester Exam 120 Minutes CSL302/612 Artificial Intelligence End-Semester Exam 120 Minutes Name: Roll Number: Please read the following instructions carefully Ø Calculators are allowed. However, laptops or mobile phones are not

More information

A Probabilistic Relational Model for Characterizing Situations in Dynamic Multi-Agent Systems

A Probabilistic Relational Model for Characterizing Situations in Dynamic Multi-Agent Systems A Probabilistic Relational Model for Characterizing Situations in Dynamic Multi-Agent Systems Daniel Meyer-Delius 1, Christian Plagemann 1, Georg von Wichert 2, Wendelin Feiten 2, Gisbert Lawitzky 2, and

More information

Markov Model. Model representing the different resident states of a system, and the transitions between the different states

Markov Model. Model representing the different resident states of a system, and the transitions between the different states Markov Model Model representing the different resident states of a system, and the transitions between the different states (applicable to repairable, as well as non-repairable systems) System behavior

More information

Quantitative evaluation of concurrent systems with non-markovian temporal parameters

Quantitative evaluation of concurrent systems with non-markovian temporal parameters Quantitative evaluation of concurrent systems with non-markovian temporal parameters Enrico Vicario Lab. of Software and Data Science Dept. of Information Engineering, University of Florence, Italy int.

More information

5. Introduction to Euclid s Geometry

5. Introduction to Euclid s Geometry 5. Introduction to Euclid s Geometry Multiple Choice Questions CBSE TREND SETTER PAPER _ 0 EXERCISE 5.. If the point P lies in between M and N and C is mid-point of MP, then : (A) MC + PN = MN (B) MP +

More information

Logic, Knowledge Representation and Bayesian Decision Theory

Logic, Knowledge Representation and Bayesian Decision Theory Logic, Knowledge Representation and Bayesian Decision Theory David Poole University of British Columbia Overview Knowledge representation, logic, decision theory. Belief networks Independent Choice Logic

More information

A Hybrid Approach to Recognising Activities of Daily Living from Object Use in the Home Environment

A Hybrid Approach to Recognising Activities of Daily Living from Object Use in the Home Environment informatics Article A Hybrid Approach to Recognising Activities of Daily Living from Object Use in the Home Environment Isibor Kennedy Ihianle 1, * ID, Usman Naeem 1, Syed Islam 1 and Abdel-Rahman Tawil

More information

Contents Preface The Exponential Distribution and the Poisson Process Introduction to Renewal Theory

Contents Preface The Exponential Distribution and the Poisson Process Introduction to Renewal Theory Contents Preface... v 1 The Exponential Distribution and the Poisson Process... 1 1.1 Introduction... 1 1.2 The Density, the Distribution, the Tail, and the Hazard Functions... 2 1.2.1 The Hazard Function

More information

Semi-Markovian User State Estimation and Policy Optimization for Energy Efficient Mobile Sensing

Semi-Markovian User State Estimation and Policy Optimization for Energy Efficient Mobile Sensing Semi-Markovian User State Estimation Policy Optimization for Energy Efficient Mobile Sensing ABSTRACT User context monitoring using sensors on mobile devices benefits end-users unobtrusively in real-time

More information

ONR Mine Warfare Autonomy Virtual Program Review September 7, 2017

ONR Mine Warfare Autonomy Virtual Program Review September 7, 2017 ONR Mine Warfare Autonomy Virtual Program Review September 7, 2017 Information-driven Guidance and Control for Adaptive Target Detection and Classification Silvia Ferrari Pingping Zhu and Bo Fu Mechanical

More information

Solving Zero-Sum Extensive-Form Games. Branislav Bošanský AE4M36MAS, Fall 2013, Lecture 6

Solving Zero-Sum Extensive-Form Games. Branislav Bošanský AE4M36MAS, Fall 2013, Lecture 6 Solving Zero-Sum Extensive-Form Games ranislav ošanský E4M36MS, Fall 2013, Lecture 6 Imperfect Information EFGs States Players 1 2 Information Set ctions Utility Solving II Zero-Sum EFG with perfect recall

More information

Final Exam December 12, 2017

Final Exam December 12, 2017 Introduction to Artificial Intelligence CSE 473, Autumn 2017 Dieter Fox Final Exam December 12, 2017 Directions This exam has 7 problems with 111 points shown in the table below, and you have 110 minutes

More information

Neural Networks. Bishop PRML Ch. 5. Alireza Ghane. Feed-forward Networks Network Training Error Backpropagation Applications

Neural Networks. Bishop PRML Ch. 5. Alireza Ghane. Feed-forward Networks Network Training Error Backpropagation Applications Neural Networks Bishop PRML Ch. 5 Alireza Ghane Neural Networks Alireza Ghane / Greg Mori 1 Neural Networks Neural networks arise from attempts to model human/animal brains Many models, many claims of

More information

Process Scheduling for RTS. RTS Scheduling Approach. Cyclic Executive Approach

Process Scheduling for RTS. RTS Scheduling Approach. Cyclic Executive Approach Process Scheduling for RTS Dr. Hugh Melvin, Dept. of IT, NUI,G RTS Scheduling Approach RTS typically control multiple parameters concurrently Eg. Flight Control System Speed, altitude, inclination etc..

More information

Integrated Electricity Demand and Price Forecasting

Integrated Electricity Demand and Price Forecasting Integrated Electricity Demand and Price Forecasting Create and Evaluate Forecasting Models The many interrelated factors which influence demand for electricity cannot be directly modeled by closed-form

More information

Modeling Data with Linear Combinations of Basis Functions. Read Chapter 3 in the text by Bishop

Modeling Data with Linear Combinations of Basis Functions. Read Chapter 3 in the text by Bishop Modeling Data with Linear Combinations of Basis Functions Read Chapter 3 in the text by Bishop A Type of Supervised Learning Problem We want to model data (x 1, t 1 ),..., (x N, t N ), where x i is a vector

More information

Activity Mining in Sensor Networks

Activity Mining in Sensor Networks MITSUBISHI ELECTRIC RESEARCH LABORATORIES http://www.merl.com Activity Mining in Sensor Networks Christopher R. Wren, David C. Minnen TR2004-135 December 2004 Abstract We present results from the exploration

More information

Class 11 Non-Parametric Models of a Service System; GI/GI/1, GI/GI/n: Exact & Approximate Analysis.

Class 11 Non-Parametric Models of a Service System; GI/GI/1, GI/GI/n: Exact & Approximate Analysis. Service Engineering Class 11 Non-Parametric Models of a Service System; GI/GI/1, GI/GI/n: Exact & Approximate Analysis. G/G/1 Queue: Virtual Waiting Time (Unfinished Work). GI/GI/1: Lindley s Equations

More information

Multistate Modeling and Applications

Multistate Modeling and Applications Multistate Modeling and Applications Yang Yang Department of Statistics University of Michigan, Ann Arbor IBM Research Graduate Student Workshop: Statistics for a Smarter Planet Yang Yang (UM, Ann Arbor)

More information

Spatial Transformation

Spatial Transformation Spatial Transformation Presented by Liqun Chen June 30, 2017 1 Overview 2 Spatial Transformer Networks 3 STN experiments 4 Recurrent Models of Visual Attention (RAM) 5 Recurrent Models of Visual Attention

More information

As Soon As Probable. O. Maler, J.-F. Kempf, M. Bozga. March 15, VERIMAG Grenoble, France

As Soon As Probable. O. Maler, J.-F. Kempf, M. Bozga. March 15, VERIMAG Grenoble, France As Soon As Probable O. Maler, J.-F. Kempf, M. Bozga VERIMAG Grenoble, France March 15, 2013 O. Maler, J.-F. Kempf, M. Bozga (VERIMAG Grenoble, France) As Soon As Probable March 15, 2013 1 / 42 Executive

More information

Forecasting Wind Ramps

Forecasting Wind Ramps Forecasting Wind Ramps Erin Summers and Anand Subramanian Jan 5, 20 Introduction The recent increase in the number of wind power producers has necessitated changes in the methods power system operators

More information

UNIT Define joint distribution and joint probability density function for the two random variables X and Y.

UNIT Define joint distribution and joint probability density function for the two random variables X and Y. UNIT 4 1. Define joint distribution and joint probability density function for the two random variables X and Y. Let and represent the probability distribution functions of two random variables X and Y

More information

A Bayesian Perspective on Residential Demand Response Using Smart Meter Data

A Bayesian Perspective on Residential Demand Response Using Smart Meter Data A Bayesian Perspective on Residential Demand Response Using Smart Meter Data Datong-Paul Zhou, Maximilian Balandat, and Claire Tomlin University of California, Berkeley [datong.zhou, balandat, tomlin]@eecs.berkeley.edu

More information

27 : Distributed Monte Carlo Markov Chain. 1 Recap of MCMC and Naive Parallel Gibbs Sampling

27 : Distributed Monte Carlo Markov Chain. 1 Recap of MCMC and Naive Parallel Gibbs Sampling 10-708: Probabilistic Graphical Models 10-708, Spring 2014 27 : Distributed Monte Carlo Markov Chain Lecturer: Eric P. Xing Scribes: Pengtao Xie, Khoa Luu In this scribe, we are going to review the Parallel

More information

Final Exam December 12, 2017

Final Exam December 12, 2017 Introduction to Artificial Intelligence CSE 473, Autumn 2017 Dieter Fox Final Exam December 12, 2017 Directions This exam has 7 problems with 111 points shown in the table below, and you have 110 minutes

More information

COMS 4771 Lecture Boosting 1 / 16

COMS 4771 Lecture Boosting 1 / 16 COMS 4771 Lecture 12 1. Boosting 1 / 16 Boosting What is boosting? Boosting: Using a learning algorithm that provides rough rules-of-thumb to construct a very accurate predictor. 3 / 16 What is boosting?

More information

Feed-forward Networks Network Training Error Backpropagation Applications. Neural Networks. Oliver Schulte - CMPT 726. Bishop PRML Ch.

Feed-forward Networks Network Training Error Backpropagation Applications. Neural Networks. Oliver Schulte - CMPT 726. Bishop PRML Ch. Neural Networks Oliver Schulte - CMPT 726 Bishop PRML Ch. 5 Neural Networks Neural networks arise from attempts to model human/animal brains Many models, many claims of biological plausibility We will

More information

Introduction to Restricted Boltzmann Machines

Introduction to Restricted Boltzmann Machines Introduction to Restricted Boltzmann Machines Ilija Bogunovic and Edo Collins EPFL {ilija.bogunovic,edo.collins}@epfl.ch October 13, 2014 Introduction Ingredients: 1. Probabilistic graphical models (undirected,

More information

6 Reinforcement Learning

6 Reinforcement Learning 6 Reinforcement Learning As discussed above, a basic form of supervised learning is function approximation, relating input vectors to output vectors, or, more generally, finding density functions p(y,

More information

CptS 570 Machine Learning School of EECS Washington State University. CptS Machine Learning 1

CptS 570 Machine Learning School of EECS Washington State University. CptS Machine Learning 1 CptS 570 Machine Learning School of EECS Washington State University CptS 570 - Machine Learning 1 IEEE Expert, October 1996 CptS 570 - Machine Learning 2 Given sample S from all possible examples D Learner

More information

A DARK GREY P O N T, with a Switch Tail, and a small Star on the Forehead. Any

A DARK GREY P O N T, with a Switch Tail, and a small Star on the Forehead. Any Y Y Y X X «/ YY Y Y ««Y x ) & \ & & } # Y \#$& / Y Y X» \\ / X X X x & Y Y X «q «z \x» = q Y # % \ & [ & Z \ & { + % ) / / «q zy» / & / / / & x x X / % % ) Y x X Y $ Z % Y Y x x } / % «] «] # z» & Y X»

More information

Glossary availability cellular manufacturing closed queueing network coefficient of variation (CV) conditional probability CONWIP

Glossary availability cellular manufacturing closed queueing network coefficient of variation (CV) conditional probability CONWIP Glossary availability The long-run average fraction of time that the processor is available for processing jobs, denoted by a (p. 113). cellular manufacturing The concept of organizing the factory into

More information

Uncertainty Quantification in Performance Evaluation of Manufacturing Processes

Uncertainty Quantification in Performance Evaluation of Manufacturing Processes Uncertainty Quantification in Performance Evaluation of Manufacturing Processes Manufacturing Systems October 27, 2014 Saideep Nannapaneni, Sankaran Mahadevan Vanderbilt University, Nashville, TN Acknowledgement

More information

Empirical Risk Minimization Algorithms

Empirical Risk Minimization Algorithms Empirical Risk Minimization Algorithms Tirgul 2 Part I November 2016 Reminder Domain set, X : the set of objects that we wish to label. Label set, Y : the set of possible labels. A prediction rule, h:

More information

STOCHASTIC MODELS FOR RELIABILITY, AVAILABILITY, AND MAINTAINABILITY

STOCHASTIC MODELS FOR RELIABILITY, AVAILABILITY, AND MAINTAINABILITY STOCHASTIC MODELS FOR RELIABILITY, AVAILABILITY, AND MAINTAINABILITY Ph.D. Assistant Professor Industrial and Systems Engineering Auburn University RAM IX Summit November 2 nd 2016 Outline Introduction

More information

Basic concepts of probability theory

Basic concepts of probability theory Basic concepts of probability theory Random variable discrete/continuous random variable Transform Z transform, Laplace transform Distribution Geometric, mixed-geometric, Binomial, Poisson, exponential,

More information

Learning Methods for Linear Detectors

Learning Methods for Linear Detectors Intelligent Systems: Reasoning and Recognition James L. Crowley ENSIMAG 2 / MoSIG M1 Second Semester 2011/2012 Lesson 20 27 April 2012 Contents Learning Methods for Linear Detectors Learning Linear Detectors...2

More information

Predicting Target Events in Industrial Domains

Predicting Target Events in Industrial Domains Predicting Target Events in Industrial Domains Julio Borges 1(B), Martin A. Neumann 1, Christian Bauer 2,YongDing 1, Till Riedel 1, and Michael Beigl 1 1 TECO, Karlsruhe Institute of Technology (KIT),

More information

Partially Observable Markov Decision Processes (POMDPs)

Partially Observable Markov Decision Processes (POMDPs) Partially Observable Markov Decision Processes (POMDPs) Sachin Patil Guest Lecture: CS287 Advanced Robotics Slides adapted from Pieter Abbeel, Alex Lee Outline Introduction to POMDPs Locally Optimal Solutions

More information

IN THIS paper we investigate the diagnosability of stochastic

IN THIS paper we investigate the diagnosability of stochastic 476 IEEE TRANSACTIONS ON AUTOMATIC CONTROL, VOL 50, NO 4, APRIL 2005 Diagnosability of Stochastic Discrete-Event Systems David Thorsley and Demosthenis Teneketzis, Fellow, IEEE Abstract We investigate

More information

Abstract. 1. Introduction

Abstract. 1. Introduction Abstract Repairable system reliability: recent developments in CBM optimization A.K.S. Jardine, D. Banjevic, N. Montgomery, A. Pak Department of Mechanical and Industrial Engineering, University of Toronto,

More information

Fair Divsion in Theory and Practice

Fair Divsion in Theory and Practice Fair Divsion in Theory and Practice Ron Cytron (Computer Science) Maggie Penn (Political Science) Lecture 6-b: Arrow s Theorem 1 Arrow s Theorem The general question: Given a collection of individuals

More information

Performance Evaluation. Transient analysis of non-markovian models using stochastic state classes

Performance Evaluation. Transient analysis of non-markovian models using stochastic state classes Performance Evaluation ( ) Contents lists available at SciVerse ScienceDirect Performance Evaluation journal homepage: www.elsevier.com/locate/peva Transient analysis of non-markovian models using stochastic

More information

Introduction to Mobile Robotics Probabilistic Robotics

Introduction to Mobile Robotics Probabilistic Robotics Introduction to Mobile Robotics Probabilistic Robotics Wolfram Burgard 1 Probabilistic Robotics Key idea: Explicit representation of uncertainty (using the calculus of probability theory) Perception Action

More information

Data Analyzing and Daily Activity Learning with Hidden Markov Model

Data Analyzing and Daily Activity Learning with Hidden Markov Model Data Analyzing and Daily Activity Learning with Hidden Markov Model GuoQing Yin and Dietmar Bruckner Institute of Computer Technology Vienna University of Technology, Austria, Europe {yin, bruckner}@ict.tuwien.ac.at

More information

Bayesian Networks Inference with Probabilistic Graphical Models

Bayesian Networks Inference with Probabilistic Graphical Models 4190.408 2016-Spring Bayesian Networks Inference with Probabilistic Graphical Models Byoung-Tak Zhang intelligence Lab Seoul National University 4190.408 Artificial (2016-Spring) 1 Machine Learning? Learning

More information

Reliability Engineering I

Reliability Engineering I Happiness is taking the reliability final exam. Reliability Engineering I ENM/MSC 565 Review for the Final Exam Vital Statistics What R&M concepts covered in the course When Monday April 29 from 4:30 6:00

More information

Multimedia Databases. Previous Lecture Video Abstraction Video Abstraction Example 6/20/2013

Multimedia Databases. Previous Lecture Video Abstraction Video Abstraction Example 6/20/2013 Previous Lecture Multimedia Databases Hidden Markov Models (continued from last lecture) Introduction into Video Retrieval Wolf-Tilo Balke Silviu Homoceanu Institut für Informationssysteme Technische Universität

More information

SVMs: Non-Separable Data, Convex Surrogate Loss, Multi-Class Classification, Kernels

SVMs: Non-Separable Data, Convex Surrogate Loss, Multi-Class Classification, Kernels SVMs: Non-Separable Data, Convex Surrogate Loss, Multi-Class Classification, Kernels Karl Stratos June 21, 2018 1 / 33 Tangent: Some Loose Ends in Logistic Regression Polynomial feature expansion in logistic

More information

University of Genova - DITEN. Smart Patrolling. video and SIgnal Processing for Telecommunications ISIP40

University of Genova - DITEN. Smart Patrolling. video and SIgnal Processing for Telecommunications ISIP40 University of Genova - DITEN Smart Patrolling 1 Smart Patrolling Detection of the intruder Tracking of the intruder A cognitive node will active an operator, describing on his mobile terminal the characteristic

More information

Methods for the specification and verification of business processes MPB (6 cfu, 295AA)

Methods for the specification and verification of business processes MPB (6 cfu, 295AA) Methods for the specification and verification of business processes MPB (6 cfu, 295AA) Roberto Bruni http://www.di.unipi.it/~bruni 17 - Diagnosis for WF nets 1 Object We study suitable diagnosis techniques

More information

Kalman filtering and friends: Inference in time series models. Herke van Hoof slides mostly by Michael Rubinstein

Kalman filtering and friends: Inference in time series models. Herke van Hoof slides mostly by Michael Rubinstein Kalman filtering and friends: Inference in time series models Herke van Hoof slides mostly by Michael Rubinstein Problem overview Goal Estimate most probable state at time k using measurement up to time

More information

2. Stochastic Time Petri Nets

2. Stochastic Time Petri Nets 316 A. Horváth et al. / Performance Evaluation 69 (2012) 315 335 kernels can be expressed in closed-form in terms of the exponential of the matrix describing the subordinated CTMC [8] and evaluated numerically

More information

Intro to Probability. Andrei Barbu

Intro to Probability. Andrei Barbu Intro to Probability Andrei Barbu Some problems Some problems A means to capture uncertainty Some problems A means to capture uncertainty You have data from two sources, are they different? Some problems

More information

Two-Sample Inferential Statistics

Two-Sample Inferential Statistics The t Test for Two Independent Samples 1 Two-Sample Inferential Statistics In an experiment there are two or more conditions One condition is often called the control condition in which the treatment is

More information

NCDREC: A Decomposability Inspired Framework for Top-N Recommendation

NCDREC: A Decomposability Inspired Framework for Top-N Recommendation NCDREC: A Decomposability Inspired Framework for Top-N Recommendation Athanasios N. Nikolakopoulos,2 John D. Garofalakis,2 Computer Engineering and Informatics Department, University of Patras, Greece

More information

Proceedings of the 2017 Winter Simulation Conference W. K. V. Chan, A. D Ambrogio, G. Zacharewicz, N. Mustafee, G. Wainer, and E. Page, eds.

Proceedings of the 2017 Winter Simulation Conference W. K. V. Chan, A. D Ambrogio, G. Zacharewicz, N. Mustafee, G. Wainer, and E. Page, eds. Proceedings of the 2017 Winter Simulation Conference W. K. V. Chan, A. D Ambrogio, G. Zacharewicz, N. Mustafee, G. Wainer, and E. Page, eds. SIMULATION-BASED PREDICTIVE ANALYTICS FOR DYNAMIC QUEUEING SYSTEMS

More information

Contents Lecture 4. Lecture 4 Linear Discriminant Analysis. Summary of Lecture 3 (II/II) Summary of Lecture 3 (I/II)

Contents Lecture 4. Lecture 4 Linear Discriminant Analysis. Summary of Lecture 3 (II/II) Summary of Lecture 3 (I/II) Contents Lecture Lecture Linear Discriminant Analysis Fredrik Lindsten Division of Systems and Control Department of Information Technology Uppsala University Email: fredriklindsten@ituuse Summary of lecture

More information

Partially Observable Markov Decision Processes (POMDPs) Pieter Abbeel UC Berkeley EECS

Partially Observable Markov Decision Processes (POMDPs) Pieter Abbeel UC Berkeley EECS Partially Observable Markov Decision Processes (POMDPs) Pieter Abbeel UC Berkeley EECS Many slides adapted from Jur van den Berg Outline POMDPs Separation Principle / Certainty Equivalence Locally Optimal

More information

Acyclic discrete phase type distributions: properties and a parameter estimation algorithm

Acyclic discrete phase type distributions: properties and a parameter estimation algorithm Performance Evaluation 54 (2003) 1 32 Acyclic discrete phase type distributions: properties and a parameter estimation algorithm A. Bobbio a, A. Horváth b, M. Scarpa c, M. Telek b, a Dip. di Scienze e

More information

CS 347 Parallel and Distributed Data Processing

CS 347 Parallel and Distributed Data Processing CS 347 Parallel and Distributed Data Processing Spring 2016 & Clocks, Clocks, and the Ordering of Events in a Distributed System. L. Lamport, Communications of the ACM, 1978 Notes 15: & Clocks CS 347 Notes

More information

Recap. Probability, stochastic processes, Markov chains. ELEC-C7210 Modeling and analysis of communication networks

Recap. Probability, stochastic processes, Markov chains. ELEC-C7210 Modeling and analysis of communication networks Recap Probability, stochastic processes, Markov chains ELEC-C7210 Modeling and analysis of communication networks 1 Recap: Probability theory important distributions Discrete distributions Geometric distribution

More information

Homework 1/Solutions. Graded Exercises

Homework 1/Solutions. Graded Exercises MTH 310-3 Abstract Algebra I and Number Theory S18 Homework 1/Solutions Graded Exercises Exercise 1. Below are parts of the addition table and parts of the multiplication table of a ring. Complete both

More information

A Probabilistic Relational Model for Characterizing Situations in Dynamic Multi-Agent Systems

A Probabilistic Relational Model for Characterizing Situations in Dynamic Multi-Agent Systems A Probabilistic Relational Model for Characterizing Situations in Dynamic Multi-Agent Systems Daniel Meyer-Delius 1, Christian Plagemann 1, Georg von Wichert 2, Wendelin Feiten 2, Gisbert Lawitzky 2, and

More information

Lecture 3: Policy Evaluation Without Knowing How the World Works / Model Free Policy Evaluation

Lecture 3: Policy Evaluation Without Knowing How the World Works / Model Free Policy Evaluation Lecture 3: Policy Evaluation Without Knowing How the World Works / Model Free Policy Evaluation CS234: RL Emma Brunskill Winter 2018 Material builds on structure from David SIlver s Lecture 4: Model-Free

More information

MAT SYS 5120 (Winter 2012) Assignment 5 (not to be submitted) There are 4 questions.

MAT SYS 5120 (Winter 2012) Assignment 5 (not to be submitted) There are 4 questions. MAT 4371 - SYS 5120 (Winter 2012) Assignment 5 (not to be submitted) There are 4 questions. Question 1: Consider the following generator for a continuous time Markov chain. 4 1 3 Q = 2 5 3 5 2 7 (a) Give

More information

Stochastic process. X, a series of random variables indexed by t

Stochastic process. X, a series of random variables indexed by t Stochastic process X, a series of random variables indexed by t X={X(t), t 0} is a continuous time stochastic process X={X(t), t=0,1, } is a discrete time stochastic process X(t) is the state at time t,

More information

LIMITS FOR QUEUES AS THE WAITING ROOM GROWS. Bell Communications Research AT&T Bell Laboratories Red Bank, NJ Murray Hill, NJ 07974

LIMITS FOR QUEUES AS THE WAITING ROOM GROWS. Bell Communications Research AT&T Bell Laboratories Red Bank, NJ Murray Hill, NJ 07974 LIMITS FOR QUEUES AS THE WAITING ROOM GROWS by Daniel P. Heyman Ward Whitt Bell Communications Research AT&T Bell Laboratories Red Bank, NJ 07701 Murray Hill, NJ 07974 May 11, 1988 ABSTRACT We study the

More information

J. MEDHI STOCHASTIC MODELS IN QUEUEING THEORY

J. MEDHI STOCHASTIC MODELS IN QUEUEING THEORY J. MEDHI STOCHASTIC MODELS IN QUEUEING THEORY SECOND EDITION ACADEMIC PRESS An imprint of Elsevier Science Amsterdam Boston London New York Oxford Paris San Diego San Francisco Singapore Sydney Tokyo Contents

More information

A Gentle Introduction to Reinforcement Learning

A Gentle Introduction to Reinforcement Learning A Gentle Introduction to Reinforcement Learning Alexander Jung 2018 1 Introduction and Motivation Consider the cleaning robot Rumba which has to clean the office room B329. In order to keep things simple,

More information

CAP Plan, Activity, and Intent Recognition

CAP Plan, Activity, and Intent Recognition CAP6938-02 Plan, Activity, and Intent Recognition Lecture 10: Sequential Decision-Making Under Uncertainty (part 1) MDPs and POMDPs Instructor: Dr. Gita Sukthankar Email: gitars@eecs.ucf.edu SP2-1 Reminder

More information

Learning theory. Ensemble methods. Boosting. Boosting: history

Learning theory. Ensemble methods. Boosting. Boosting: history Learning theory Probability distribution P over X {0, 1}; let (X, Y ) P. We get S := {(x i, y i )} n i=1, an iid sample from P. Ensemble methods Goal: Fix ɛ, δ (0, 1). With probability at least 1 δ (over

More information

Exact Inference by Complete Enumeration

Exact Inference by Complete Enumeration 21 Exact Inference by Complete Enumeration We open our toolbox of methods for handling probabilities by discussing a brute-force inference method: complete enumeration of all hypotheses, and evaluation

More information

Switching-state Dynamical Modeling of Daily Behavioral Data

Switching-state Dynamical Modeling of Daily Behavioral Data Switching-state Dynamical Modeling of Daily Behavioral Data Randy Ardywibowo Shuai Huang Cao Xiao Shupeng Gui Yu Cheng Ji Liu Xiaoning Qian Texas A&M University University of Washington IBM T.J. Watson

More information

Lecture notes: Rust (1987) Economics : M. Shum 1

Lecture notes: Rust (1987) Economics : M. Shum 1 Economics 180.672: M. Shum 1 Estimate the parameters of a dynamic optimization problem: when to replace engine of a bus? This is a another practical example of an optimal stopping problem, which is easily

More information

A PARAMETRIC DECOMPOSITION BASED APPROACH FOR MULTI-CLASS CLOSED QUEUING NETWORKS WITH SYNCHRONIZATION STATIONS

A PARAMETRIC DECOMPOSITION BASED APPROACH FOR MULTI-CLASS CLOSED QUEUING NETWORKS WITH SYNCHRONIZATION STATIONS A PARAMETRIC DECOMPOSITION BASED APPROACH FOR MULTI-CLASS CLOSED QUEUING NETWORKS WITH SYNCHRONIZATION STATIONS Kumar Satyam and Ananth Krishnamurthy Department of Decision Sciences and Engineering Systems,

More information

Failure prognostics in a particle filtering framework Application to a PEMFC stack

Failure prognostics in a particle filtering framework Application to a PEMFC stack Failure prognostics in a particle filtering framework Application to a PEMFC stack Marine Jouin Rafael Gouriveau, Daniel Hissel, Noureddine Zerhouni, Marie-Cécile Péra FEMTO-ST Institute, UMR CNRS 6174,

More information

CS 234 Midterm - Winter

CS 234 Midterm - Winter CS 234 Midterm - Winter 2017-18 **Do not turn this page until you are instructed to do so. Instructions Please answer the following questions to the best of your ability. Read all the questions first before

More information

Lecture 18: Noise modeling and introduction to decision theory

Lecture 18: Noise modeling and introduction to decision theory Lecture 8: oise modeling and introduction to decision theory Learning Objectives: Hypothesis testing The receiver operator characteristic (ROC curve Bayes s Theorem, positive and negative predictive value

More information

State-space Model. Eduardo Rossi University of Pavia. November Rossi State-space Model Financial Econometrics / 49

State-space Model. Eduardo Rossi University of Pavia. November Rossi State-space Model Financial Econometrics / 49 State-space Model Eduardo Rossi University of Pavia November 2013 Rossi State-space Model Financial Econometrics - 2013 1 / 49 Outline 1 Introduction 2 The Kalman filter 3 Forecast errors 4 State smoothing

More information