🖍️
gitbook_docs
  • Introduction
  • Machine Learning
    • Recommended Courses
      • For Undergrad Research
      • Math for Machine Learning
    • ML Notes
      • Covariance Correlation
      • Feature Selection
      • Linear Regression
      • Entropy, Cross-Entropy, KL Divergence
      • Bayesian Classifier
        • Terminology Review
        • Bayesian Classifier for Normally Distributed classes
      • Linear Discriminant Analysis
      • Logistic Regression
        • Logistic Regression Math
      • Logistic Regression-MaximumLikelihood
      • SVM
        • SVM concept
        • SVM math
      • Cross Validation
      • Parameter, Density Estimation
        • MAP, MLE
        • Gaussian Mixture Model
      • E-M
      • Density Estimation(non-parametric)
      • Unsupervised Learning
      • Clustering
      • kNN
      • WaveletTransform
      • Decision Tree
    • Probability and Statistics for Machine Learning
      • Introduction
      • Basics of Data Analysis
      • Probability for Discrete Random Variable
      • Poisson Distribution
      • Chi-Square Distribution
      • P-value and Statistical Hypothesis
      • Power and Sample Size
      • Hypothesis Test Old
      • Hypothesis Test
      • Multi Armed Bandit
      • Bayesian Inference
      • Bayesian Updating with Continuous Priors
      • Discrete Distribution
      • Comparison of Bayesian and frequentist inference
      • Confidence Intervals for Normal Data
      • Frequenist Methods
      • Null Hypothesis Significance Testing
      • Confidence Intervals: Three Views
      • Confidence Intervals for the Mean of Non-normal Data
      • Probabilistic Prediction
  • Industrial AI
    • PHM Dataset
    • BearingFault_Journal
      • Support Vector Machine based
      • Autoregressive(AR) model based
      • Envelope Extraction based
      • Wavelet Decomposition based
      • Prediction of RUL with Deep Convolution Nueral Network
      • Prediction of RUL with Information Entropy
      • Feature Model and Feature Selection
    • TempCore Journal
      • Machine learning of mechanical properties of steels
      • Online prediction of mechanical properties of hot rolled steel plate using machine learning
      • Prediction and Analysis of Tensile Properties of Austenitic Stainless Steel Using Artificial Neural
      • Tempcore, new process for the production of high quality reinforcing
      • TEMPCORE, the most convenient process to produce low cost high strength rebars from 8 to 75 mm
      • Experimental investigation and simulation of structure and tensile properties of Tempcore treated re
    • Notes
  • LiDAR
    • Processing of Point Cloud
    • Intro. 3D Object Detection
    • PointNet
    • PointNet++
    • Frustrum-PointNet
    • VoxelNet
    • Point RCNN
    • PointPillars
    • LaserNet
  • Simulator
    • Simulator List
    • CARLA
    • Airsim
      • Setup
      • Tutorial
        • T#1
        • T#2
        • T#3: Opencv CPP
        • T#4: Opencv Py
        • Untitled
        • T#5: End2End Driving
  • Resources
    • Useful Resources
    • Github
    • Jekyll
  • Reinforcement Learning
    • RL Overview
      • RL Bootcamp
      • MIT Deep RL
    • Textbook
    • Basics
    • Continuous Space RL
  • Unsupervised Learning
    • Introduction
  • Unclassified
    • Ethics
    • Conference Guideline
  • FPGA
    • Untitled
  • Numerical Method
    • NM API reference
Powered by GitBook
On this page
  • Theory
  • Poisson Distribution 이란?
  • PMF of Poisson Distribution
  • When use?
  • Poisson Paradigm (Poisson Approximation)
  • 이항분포가 어떻게 푸아송 분포로 수렴하려는가?
  • Practice (MATLAB)
  • 푸아송 분포의 pdf 계산하기
  • 푸아송 분포의 cdf 계산하기
  • 푸아송 분포 pdf와 정규분포 pdf 비교하기

Was this helpful?

  1. Machine Learning
  2. Probability and Statistics for Machine Learning

Poisson Distribution

PreviousProbability for Discrete Random VariableNextChi-Square Distribution

Last updated 3 years ago

Was this helpful?

Theory

Ref)

Poisson Distribution 이란?

  • 표집된 단위 시간 (혹은 단위 공간)에서 발생한 사건의 도수 분포

  • 단위 시간 안에 몇 번 발생할 것인지를 표현하는 이산 확률 분포

  • 푸아송 분포의 기대값 및 분산은 λ

f(n,λ)=λne−λn!f(n, \lambda)= {\lambda^ne^{-\lambda} \over n!}f(n,λ)=n!λne−λ​

PMF of Poisson Distribution

f(X=k)=λke−λk!      k∈{0,1,2⋯ }f(X = k) = {{{\lambda ^k}{e^{ - \lambda }}} \over {k!}}\,\,\,\,\,\,k \in \{ 0,1,2 \cdots \}f(X=k)=k!λke−λ​k∈{0,1,2⋯}
  • Is it valid PMF?

∑k=0∞λke−λk! =e−λeλ=1\sum\limits_{k = 0}^\infty {{{{\lambda ^k}{e^{ - \lambda }}} \over {k!}}\,} = {e^{ - \lambda }}{e^\lambda } = 1k=0∑∞​k!λke−λ​=e−λeλ=1
  • Expectation value

\eqalign{ & E(X) = {e^{ - \lambda }}\sum\limits_{k = 0}^\infty {{{k{\lambda ^k}} \over {k!}}} \cr & \,\,\,\,\,\,\,\,\,\,\,\,\,\, = {e^{ - \lambda }}\sum\limits_{k = 1}^\infty {{{{\lambda ^{k - 1}}\lambda } \over {(k - 1)!}}} = \lambda \cr}

When use?

  • 주어진 시간,거리,면적 등에서 임의 사건이 발생하는 횟수를 세야 하는 경우에 적합.

    • Ex) 초당 클릭 횟수, 시간당 매장에 들어오는 사람 수, 1분에 네트워크에서 손실되는 패킷 수 등

  • 굉장히 여러 번 시행하지만 성공확률이 낮은 경우에 사용.

    1. Number of emails in hour.

    2. Number of chips in chocolate chip cookies.

    3. Number of earthquakes in a year in some region.

Poisson Paradigm (Poisson Approximation)

  • Events A~1~, A~2~, ... A~n~, P(A~j~)=p~j~ (n is large, p~j~ is small)\

  • 각각의 사건들이 "independent" or "weakly independent" 하다면,

    Then # of A~j~'s that occur is approximated,

    Pois(λ);    λ=∑j=1npjPois(\lambda );\,\,\,\,\lambda = \sum\limits_{j = 1}^n {{p_j}}Pois(λ);λ=j=1∑n​pj​

이항분포가 어떻게 푸아송 분포로 수렴하려는가?

\eqalign{ & X \sim Bin(n,p),\,\,\,let\,\,n \to \infty ,\,\,p \to 0,\,\,\lambda = np\,{\rm{is}}\,{\rm{held}}\,{\rm{constant}}{\rm{.}} \cr & {\rm{Find}}\,{\rm{what}}\,{\rm{happens}}\,{\rm{to}}\,P(X = k) = \left( \matrix{ n \hfill \cr k \hfill \cr} \right){p^k}{(1 - p)^{n - k}},\,\,\,\,k\,{\rm{fixed}}{\rm{.}} \cr & \,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\, = {{n(n - 1) \cdots (n - k + 1)} \over {k!}}{{{\lambda ^k}} \over {{n^k}}}{\left( {1 - {\lambda \over n}} \right)^n}{\left( {1 - {\lambda \over n}} \right)^{ - k}} \cr & \,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\, = {{{\lambda ^k}} \over {k!}}{e^{ - \lambda }},\,\,\,\,Poisson\,PMF\,at\,k \cr}

Practice (MATLAB)

푸아송 분포의 pdf 계산하기

모수 lambda=4 를 갖는 푸아송 분포의 pdf를 계산해보자.

x = 0:15;
y = poisspdf(x, 4);

figure();
bar(x,y,1)
xlabel('Observation')
ylabel('Probability')

푸아송 분포의 cdf 계산하기

x = 0:15;
y = poisscdf(x,4);

figure;
stairs(x,y)
xlabel('Observation')
ylabel('Cumulative Probability')

푸아송 분포 pdf와 정규분포 pdf 비교하기

labmda가 크면 푸아송 분포는 평균 lambda와 분산 lambda를 갖는 정규분포로 근사가 가능하다.

모수 labmda=50을 갖는 푸아송 분포의 pdf를 계산해보자.

lambda = 50;
x1 = 0:100;
y1 = poisspdf(x1, lambda);

mu = lambda;
sigma = sqrt(lambda);
x2 = 0:0.1:100;
y2 = normpdf(x2,mu,sigma);

figure;
bar(x1, y1, 1)
hold on
plot(x2, y2, 'LineWidth', 2)
xlabel('Observation')
ylabel('Probability')
title('Poisson and Normal pdfs')
legend('Poisson Distribution', 'Normal distribution', 'location', 'northwest')
hold off

Ref)

Harvard Statistics110 - The Poisson distribution
MATLAB Poisson distribution
image-20210413200629962
image-20210413200713361
image-20210413201247179