Scribd Logo
Upload

Welcome to Scribd!

  • 149 views

    Cse 150 HW1

    Uploaded by

    n0nce
    This document outlines an assignment for a probabilistic reasoning course. It includes 4 problems: 1) Defining and analyzing the Kullback-Leibler distance metric for comparing probability distributions. 2) Showing equivalence between statements of conditional independence between random variables. 3) Providing examples that illustrate patterns of commonsense probabilistic reasoning like explaining away and accumulating evidence. 4) Computing probabilities in a Bayesian network model about earthquakes, burglaries, and alarm systems to practice probabilistic inference.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Cse 150 HW1

    Uploaded by

    n0nce
    149 views2 pages
    This document outlines an assignment for a probabilistic reasoning course. It includes 4 problems: 1) Defining and analyzing the Kullback-Leibler distance metric for comparing probability distributions. 2) Showing equivalence between statements of conditional independence between random variables. 3) Providing examples that illustrate patterns of commonsense probabilistic reasoning like explaining away and accumulating evidence. 4) Computing probabilities in a Bayesian network model about earthquakes, burglaries, and alarm systems to practice probabilistic inference.

    Original Description:

    CSE 150 HW1.

    Original Title

    CSE 150 HW1

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document outlines an assignment for a probabilistic reasoning course. It includes 4 problems: 1) Defining and analyzing the Kullback-Leibler distance metric for comparing probability distributions. 2) Showing equivalence between statements of conditional independence between random variables. 3) Providing examples that illustrate patterns of commonsense probabilistic reasoning like explaining away and accumulating evidence. 4) Computing probabilities in a Bayesian network model about earthquakes, burglaries, and alarm systems to practice probabilistic inference.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    149 views2 pages

    Cse 150 HW1

    Uploaded by

    n0nce
    This document outlines an assignment for a probabilistic reasoning course. It includes 4 problems: 1) Defining and analyzing the Kullback-Leibler distance metric for comparing probability distributions. 2) Showing equivalence between statements of conditional independence between random variables. 3) Providing examples that illustrate patterns of commonsense probabilistic reasoning like explaining away and accumulating evidence. 4) Computing probabilities in a Bayesian network model about earthquakes, burglaries, and alarm systems to practice probabilistic inference.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 2

    CSE 150.

    Assignment 1

    Winter 2015

    Out: Tue Jan 13


    Due: Tue Jan 20 (in class)
    Supplementary reading: RN, Ch 13; KN, Ch 1.

    1.1

    Kullback-Leibler distance

    Often it is useful to measure the difference between two probability distributions over the same random
    variable. For example, as shorthand let
    pi = P (X = xi |E),
    qi = P (X = xi |E 0 )
    denote the conditional distributions over the random variable X for different pieces of evidence E 6= E 0 .
    The Kullback-Leibler (KL) distance between these distributions is defined as:
    KL(p, q) =

    pi log(pi /qi ).

    (a) By sketching graphs of log x and x 1, verify the inequality


    log x x 1,
    with equality if and only if x = 1. Confirm this result by differentiation of log x (x1). (Note: all
    logarithms in this problem are natural logarithms.)
    (b) Use the previous result to prove that KL(p, q) 0, with equality if and only if the two distributions
    pi and qi are equal.
    (c) Provide a counterexample to show that the KL distance is not a symmetric function of its arguments:
    KL(p, q) 6= KL(q, p).
    Despite this asymmetry, it is still common to refer to KL(p, q) as a measure of distance between
    probability distributions.

    1.2

    Conditional independence

    Show that the following three statements about random variables X, Y , and Z are equivalent:
    P (X, Y |Z) = P (X|Z)P (Y |Z)
    P (X|Y, Z) = P (X|Z)
    P (Y |X, Z) = P (Y |Z)
    In other words, show that each one of these statements implies the other two. You should become fluent
    with all these ways of expressing that X is conditionally independent of Y given Z.

    1.3

    Creative writing

    Attach events to the binary random variables X, Y , and Z that are consistent with the following patterns of
    commonsense reasoning. You may use different events for the different parts of the problem.
    (a) Explaining away:
    P (X = 1|Y = 1) > P (X = 1),
    P (X = 1|Y = 1, Z = 1) < P (X = 1|Y = 1)
    (b) Accumulating evidence:
    P (X = 1) > P (X = 1|Y = 1) > P (X = 1|Y = 1, Z = 1)
    (c) Conditional independence:
    P (X = 1, Y = 1) > P (X = 1)P (Y = 1)
    P (X = 1, Y = 1|Z = 1) = P (X = 1|Z = 1)P (Y = 1|Z = 1)

    1.4

    Probabilistic inference

    Recall the probabilistic model that we described in class for the binary random variables {E = Earthquake,
    B = Burglary, A = Alarm, J = JohnCalls, M = MaryCalls}. We also expressed this model as a belief
    network, with the directed acyclic graph (DAG) and conditional probability tables (CPTs) shown below:
    P(E=1) = 0.002

    Earthquake

    Burglar

    Alarm

    P(J=1|A=0) = 0.05
    P(J=1|A=1) = 0.90

    John Calls

    P(B=1) = 0.001

    P(A=1|E=0,B=0) = 0.001
    P(A=1|E=0,B=1) = 0.94
    P(A=1|E=1,B=0) = 0.29
    P(A=1|E=1,B=1) = 0.95

    Mary Calls

    P(M=1|A=0) = 0.01
    P(M=1|A=1) = 0.70

    Compute numeric values for the following probabilities, exploiting relations of conditional independence as
    much as possible to simplify your calculations. You may re-use numerical results from lecture, but otherwise
    show your work. Be careful not to drop significant digits in your answer.
    (a) P (E = 1|A = 1)
    (b) P (E = 1|A = 1, B = 1)

    (c) P (A = 1|J = 0)
    (d) P (A = 1|J = 0, M = 0)

    (e) P (A = 1|M = 1)
    (f) P (A = 1|M = 1, E = 0)

    Consider your results in (b) versus (a), (d) versus (c), and (f) versus (e). Do they seem consistent with
    commonsense patterns of reasoning?

    You might also like

    pFad - Phonifier reborn

    Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

    Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


    Alternative Proxies:

    Alternative Proxy

    pFad Proxy

    pFad v3 Proxy

    pFad v4 Proxy