Enviar búsqueda
Cargar
Gaussian Bayes Classifiers
•
1 recomendación
•
736 vistas
G
guestfee8698
Seguir
Tecnología
Educación
Denunciar
Compartir
Denunciar
Compartir
1 de 33
Descargar ahora
Descargar para leer sin conexión
Recomendados
Naive Bayes
Naive Bayes
CloudxLab
Destek vektör makineleri
Destek vektör makineleri
ozgur_dolgun
Lecture 6: Ensemble Methods
Lecture 6: Ensemble Methods
Marina Santini
Naive Bayes Classifier
Naive Bayes Classifier
Arunabha Saha
Generative Adversarial Networks
Generative Adversarial Networks
Mustafa Yagmur
Unsupervised learning clustering
Unsupervised learning clustering
Arshad Farhad
Feature Engineering in Machine Learning
Feature Engineering in Machine Learning
Knoldus Inc.
Basic Generative Adversarial Networks
Basic Generative Adversarial Networks
Dong Heon Cho
Recomendados
Naive Bayes
Naive Bayes
CloudxLab
Destek vektör makineleri
Destek vektör makineleri
ozgur_dolgun
Lecture 6: Ensemble Methods
Lecture 6: Ensemble Methods
Marina Santini
Naive Bayes Classifier
Naive Bayes Classifier
Arunabha Saha
Generative Adversarial Networks
Generative Adversarial Networks
Mustafa Yagmur
Unsupervised learning clustering
Unsupervised learning clustering
Arshad Farhad
Feature Engineering in Machine Learning
Feature Engineering in Machine Learning
Knoldus Inc.
Basic Generative Adversarial Networks
Basic Generative Adversarial Networks
Dong Heon Cho
DBSCAN : A Clustering Algorithm
DBSCAN : A Clustering Algorithm
Pınar Yahşi
Stacking ensemble
Stacking ensemble
kalung0313
Neural Networks: Rosenblatt's Perceptron
Neural Networks: Rosenblatt's Perceptron
Mostafa G. M. Mostafa
Ensemble methods in machine learning
Ensemble methods in machine learning
SANTHOSH RAJA M G
A survey of deep learning approaches to medical applications
A survey of deep learning approaches to medical applications
Joseph Paul Cohen PhD
Random Forest
Random Forest
Abdullah al Mamun
Random forest
Random forest
Ujjawal
Introduction to Machine Learning with SciKit-Learn
Introduction to Machine Learning with SciKit-Learn
Benjamin Bengfort
Data Augmentation
Data Augmentation
Md Tajul Islam
Random forest
Random forest
Musa Hawamdah
Robustness in deep learning
Robustness in deep learning
Ganesan Narayanasamy
Brain tumor classification using artificial neural network on mri images
Brain tumor classification using artificial neural network on mri images
eSAT Journals
Probability & Information theory
Probability & Information theory
성재 최
Market oriented Cloud Computing
Market oriented Cloud Computing
Jithin Parakka
Bayesian networks
Bayesian networks
Orochi Krizalid
Machine Learning Classifiers
Machine Learning Classifiers
Mostafa
Kernel Method
Kernel Method
수철 박
SVM Tutorial
SVM Tutorial
butest
KNN
KNN
West Virginia University
GMM
GMM
atulnitrkl
Short Overview of Bayes Nets
Short Overview of Bayes Nets
guestfee8698
Hope VI
Hope VI
Daniel Davis
Más contenido relacionado
La actualidad más candente
DBSCAN : A Clustering Algorithm
DBSCAN : A Clustering Algorithm
Pınar Yahşi
Stacking ensemble
Stacking ensemble
kalung0313
Neural Networks: Rosenblatt's Perceptron
Neural Networks: Rosenblatt's Perceptron
Mostafa G. M. Mostafa
Ensemble methods in machine learning
Ensemble methods in machine learning
SANTHOSH RAJA M G
A survey of deep learning approaches to medical applications
A survey of deep learning approaches to medical applications
Joseph Paul Cohen PhD
Random Forest
Random Forest
Abdullah al Mamun
Random forest
Random forest
Ujjawal
Introduction to Machine Learning with SciKit-Learn
Introduction to Machine Learning with SciKit-Learn
Benjamin Bengfort
Data Augmentation
Data Augmentation
Md Tajul Islam
Random forest
Random forest
Musa Hawamdah
Robustness in deep learning
Robustness in deep learning
Ganesan Narayanasamy
Brain tumor classification using artificial neural network on mri images
Brain tumor classification using artificial neural network on mri images
eSAT Journals
Probability & Information theory
Probability & Information theory
성재 최
Market oriented Cloud Computing
Market oriented Cloud Computing
Jithin Parakka
Bayesian networks
Bayesian networks
Orochi Krizalid
Machine Learning Classifiers
Machine Learning Classifiers
Mostafa
Kernel Method
Kernel Method
수철 박
SVM Tutorial
SVM Tutorial
butest
KNN
KNN
West Virginia University
GMM
GMM
atulnitrkl
La actualidad más candente
(20)
DBSCAN : A Clustering Algorithm
DBSCAN : A Clustering Algorithm
Stacking ensemble
Stacking ensemble
Neural Networks: Rosenblatt's Perceptron
Neural Networks: Rosenblatt's Perceptron
Ensemble methods in machine learning
Ensemble methods in machine learning
A survey of deep learning approaches to medical applications
A survey of deep learning approaches to medical applications
Random Forest
Random Forest
Random forest
Random forest
Introduction to Machine Learning with SciKit-Learn
Introduction to Machine Learning with SciKit-Learn
Data Augmentation
Data Augmentation
Random forest
Random forest
Robustness in deep learning
Robustness in deep learning
Brain tumor classification using artificial neural network on mri images
Brain tumor classification using artificial neural network on mri images
Probability & Information theory
Probability & Information theory
Market oriented Cloud Computing
Market oriented Cloud Computing
Bayesian networks
Bayesian networks
Machine Learning Classifiers
Machine Learning Classifiers
Kernel Method
Kernel Method
SVM Tutorial
SVM Tutorial
KNN
KNN
GMM
GMM
Destacado
Short Overview of Bayes Nets
Short Overview of Bayes Nets
guestfee8698
Hope VI
Hope VI
Daniel Davis
Support Vector Machines
Support Vector Machines
guestfee8698
Knowledge Exchange Team
Knowledge Exchange Team
blount_l
Metro contro Metro
Metro contro Metro
Alessandro Scardova
Ao Lua Ha Dong
Ao Lua Ha Dong
Pauldoright
Predicting Real-valued Outputs: An introduction to regression
Predicting Real-valued Outputs: An introduction to regression
guestfee8698
The Interim Advantage
The Interim Advantage
cleaman
Project AK/Informatica
Project AK/Informatica
guest391da30
Ao Lua Ha Dong
Ao Lua Ha Dong
Pauldoright
Fast Camera
Fast Camera
Pauldoright
Ao Dai Ao Ngan Yem Dao
Ao Dai Ao Ngan Yem Dao
Pauldoright
Introduzione alle app Windows Universal
Introduzione alle app Windows Universal
Alessandro Scardova
FHTM
FHTM
rjbarros
Het nieuwe werken, de mens, de cultuur en regie v11
Het nieuwe werken, de mens, de cultuur en regie v11
mscholte
Niagara falls, ny– city of missed
Niagara falls, ny– city of missed
Daniel Davis
FHTM Outline
FHTM Outline
rjbarros
Introduzione al design di Windows 10
Introduzione al design di Windows 10
Alessandro Scardova
Digital Lab Research areas
Digital Lab Research areas
blount_l
Destacado
(19)
Short Overview of Bayes Nets
Short Overview of Bayes Nets
Hope VI
Hope VI
Support Vector Machines
Support Vector Machines
Knowledge Exchange Team
Knowledge Exchange Team
Metro contro Metro
Metro contro Metro
Ao Lua Ha Dong
Ao Lua Ha Dong
Predicting Real-valued Outputs: An introduction to regression
Predicting Real-valued Outputs: An introduction to regression
The Interim Advantage
The Interim Advantage
Project AK/Informatica
Project AK/Informatica
Ao Lua Ha Dong
Ao Lua Ha Dong
Fast Camera
Fast Camera
Ao Dai Ao Ngan Yem Dao
Ao Dai Ao Ngan Yem Dao
Introduzione alle app Windows Universal
Introduzione alle app Windows Universal
FHTM
FHTM
Het nieuwe werken, de mens, de cultuur en regie v11
Het nieuwe werken, de mens, de cultuur en regie v11
Niagara falls, ny– city of missed
Niagara falls, ny– city of missed
FHTM Outline
FHTM Outline
Introduzione al design di Windows 10
Introduzione al design di Windows 10
Digital Lab Research areas
Digital Lab Research areas
Más de guestfee8698
PAC Learning
PAC Learning
guestfee8698
VC dimensio
VC dimensio
guestfee8698
Hidden Markov Models
Hidden Markov Models
guestfee8698
K-means and Hierarchical Clustering
K-means and Hierarchical Clustering
guestfee8698
Gaussian Mixture Models
Gaussian Mixture Models
guestfee8698
A Short Intro to Naive Bayesian Classifiers
A Short Intro to Naive Bayesian Classifiers
guestfee8698
Learning Bayesian Networks
Learning Bayesian Networks
guestfee8698
Inference in Bayesian Networks
Inference in Bayesian Networks
guestfee8698
Bayesian Networks
Bayesian Networks
guestfee8698
Eight Regression Algorithms
Eight Regression Algorithms
guestfee8698
Instance-based learning (aka Case-based or Memory-based or non-parametric)
Instance-based learning (aka Case-based or Memory-based or non-parametric)
guestfee8698
Neural Networks
Neural Networks
guestfee8698
Cross-Validation
Cross-Validation
guestfee8698
Maximum Likelihood Estimation
Maximum Likelihood Estimation
guestfee8698
Gaussians
Gaussians
guestfee8698
Probability for Data Miners
Probability for Data Miners
guestfee8698
Más de guestfee8698
(16)
PAC Learning
PAC Learning
VC dimensio
VC dimensio
Hidden Markov Models
Hidden Markov Models
K-means and Hierarchical Clustering
K-means and Hierarchical Clustering
Gaussian Mixture Models
Gaussian Mixture Models
A Short Intro to Naive Bayesian Classifiers
A Short Intro to Naive Bayesian Classifiers
Learning Bayesian Networks
Learning Bayesian Networks
Inference in Bayesian Networks
Inference in Bayesian Networks
Bayesian Networks
Bayesian Networks
Eight Regression Algorithms
Eight Regression Algorithms
Instance-based learning (aka Case-based or Memory-based or non-parametric)
Instance-based learning (aka Case-based or Memory-based or non-parametric)
Neural Networks
Neural Networks
Cross-Validation
Cross-Validation
Maximum Likelihood Estimation
Maximum Likelihood Estimation
Gaussians
Gaussians
Probability for Data Miners
Probability for Data Miners
Último
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
BookNet Canada
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptx
OnBoard
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
Delhi Call girls
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Igalia
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Alan Dix
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
ThousandEyes
Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen Frames
Sinan KOZAK
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
Gabriella Davis
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024
Scott Keck-Warren
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
HostedbyConfluent
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Drew Madelung
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
vu2urc
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC Architecture
Pixlogix Infotech
Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)
Allon Mureinik
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024
The Digital Insurer
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101
Paola De la Torre
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
shyamraj55
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024
Results
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
Principled Technologies
Último
(20)
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptx
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen Frames
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC Architecture
Injustice - Developers Among Us (SciFiDevCon 2024)
Injustice - Developers Among Us (SciFiDevCon 2024)
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
Gaussian Bayes Classifiers
1.
Note to other
teachers and users of these slides. Andrew would be delighted if you found this source material useful in giving you r own lectures. Feel free to use these slides verbatim, or to modify them to fit your own needs. PowerPoint originals are available. If you make use of a significant portion of these slides in your own lecture, please include this message, or the following link to the source repository of Andrew’s tutorials: http://www.cs.cmu.edu/~awm/tutorials . Comments and corrections gratefully received. Learning Gaussian Bayes Classifiers Andrew W. Moore Associate Professor School of Computer Science Carnegie Mellon University www.cs.cmu.edu/~awm awm@cs.cmu.edu 412-268-7599 Copyright © 2001, Andrew W. Moore Sep 10th, 2001 Maximum Likelihood learning of Gaussians for Classification • Why we should care • 3 seconds to teach you a new learning algorithm • What if there are 10,000 dimensions? • What if there are categorical inputs? • Examples “out the wazoo” Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 2 1
2.
Why we should
care • One of the original “Data Mining” algorithms • Very simple and effective • Demonstrates the usefulness of our earlier groundwork Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 3 Where we were at the end of the MLE lecture… Categorical Real-valued Mixed Real / inputs only inputs only Cat okay Joint BC Predict Inputs Dec Tree Classifier category Naïve BC Joint DE Gauss DE Prob- Inputs Inputs Density ability Estimator Naïve DE Predict Regressor real no. Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 4 2
3.
This lecture…
Categorical Real-valued Mixed Real / inputs only inputs only Cat okay Joint BC Gauss BC Predict Inputs Dec Tree Classifier category Naïve BC Joint DE Gauss DE Prob- Inputs Inputs Density ability Estimator Naïve DE Predict Regressor real no. Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 5 Road Map Probability Decision Trees Density PDFs Estimation Gaussians Bayes MLE Classifiers MLE of Gaussians Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 6 3
4.
Road Map
Probability Decision Trees Density PDFs Estimation Gaussians Bayes MLE Classifiers MLE of Gaussians Gaussian Bayes Classifiers Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 7 Gaussian Bayes Classifier Assumption • The i’th record in the database is created using the following algorithm 1. Generate the output (the “class”) by drawing yi~Multinomial(p1,p2,…p Ny ) 2. Generate the inputs from a Gaussian PDF that depends on the value of yi : xi ~ N(µ i ,Σ i). Test your understanding. Given Ny classes and m input attributes, how many distinct scalar parameters need to be estimated? Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 8 4
5.
MLE Gaussian Bayes
Classifier Let DB = Subset of |DB | • The i i’th record in the p mle = ------is created databasei database DB in which i using classfollowing algorithm the is y = i the output |DB| 1. Generate the output (the “class”) by drawing yi~Multinomial(p1,p2,…p Ny ) 2. Generate the inputs from a Gaussian PDF that depends on the value of yi : xi ~ N(µ i ,Σ i). Test your understanding. Given Ny classes and m input attributes, how many distinct scalar parameters need to be estimated? Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 9 MLE Gaussian Bayes Classifier Let DB = Subset of • The i i’th record in the database is created database DB in which the output class following algorithm using the is y = i 1. mle mle the output (the “class”) by Generate (µ i , Σ i )= MLE Gaussian for DBi drawing yi~Multinomial(p1,p2,…p Ny ) 2. Generate the inputs from a Gaussian PDF that depends on the value of yi : xi ~ N(µ i ,Σ i). Test your understanding. Given Ny classes and m input attributes, how many distinct scalar parameters need to be estimated? Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 10 5
6.
MLE Gaussian Bayes
Classifier Let DB = Subset of • The i i’th record in the database is created database DB in which the output class following algorithm using the is y = i 1. mle mle the output (the “class”) by Generate (µ i , Σ i )= MLE Gaussian for DBi drawing yi~Multinomial(p1,p2,…p Ny ) 2. Generate the inputs from a Gaussian PDF that depends on the value of yi : xi ~ N(µ i ,Σ i). R ( )( ) 1 ∑ R 1 ∑ x k input x k − µimle µ mle = T Test your| understanding. Given Ny classes and m− µ i attributes, how Si = xk mle mle i | DB i x k ∈DB i | DB i | x ∈DB many distinct scalar parameters need to be estimated? k i Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 11 Gaussian Bayes Classification p (x | y = i) P ( y = i) P ( y = i | x) = p( x) Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 12 6
7.
Gaussian Bayes Classification
p (x | y = i) P ( y = i) P ( y = i | x) = p( x) 1 exp − (x k − µ i ) S i (x k − µ i ) p i 1 T ( 2π ) 2 m /2 1/2 || S i || P ( y = i | x) = p (x ) How do we deal with that? Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 13 Here is a dataset age employment education edunum marital … job relation race gender hours_worked wealth country … 39 State_gov Bachelors 13 Never_married … Adm_clericalNot_in_family White Male 40 United_States poor 51 Self_emp_not_inc Bachelors 13 Married … Exec_managerial Husband White Male 13 United_States poor 39 Private HS_grad 9 Divorced … Handlers_cleaners Not_in_family White Male 40 United_States poor 54 Private 11th 7 Married … Handlers_cleaners Husband Black Male 40 United_States poor 28 Private Bachelors 13 Married … Prof_specialty Wife Black Female 40 Cuba poor 38 Private Masters 14 Married … Exec_managerial Wife White Female 40 United_States poor 50 Private 9th 5 Married_spouse_absent … Other_service Not_in_family Black Female 16 Jamaica poor 52 Self_emp_not_inc HS_grad 9 Married … Exec_managerial Husband White Male 45 United_States rich 31 Private Masters 14 Never_married … Prof_specialty Not_in_family White Female 50 United_States rich 42 Private Bachelors 13 Married … Exec_managerial Husband White Male 40 United_States rich 37 Private Some_college0 1 Married … Exec_managerial Husband Black Male 80 United_States rich 30 State_gov Bachelors 13 Married … Prof_specialty Husband Asian Male 40 India rich 24 Private Bachelors 13 Never_married … Adm_clericalOwn_child White Female 30 United_States poor 33 Private Assoc_acdm12 Never_married … Sales Not_in_family Black Male 50 United_States poor 41 Private Assoc_voc 11 Married … Craft_repairHusband Asian Male 40 *MissingValue* rich 34 Private 7th_8th 4 Married … Transport_moving Husband Amer_Indian Male 45 Mexico poor 26 Self_emp_not_inc HS_grad 9 Never_married … Farming_fishing Own_child White Male 35 United_States poor 33 Private HS_grad 9 Never_married … Machine_op_inspct White Unmarried Male 40 United_States poor 38 Private 11th 7 Married … Sales Husband White Male 50 United_States poor 44 Self_emp_not_inc Masters 14 Divorced … Exec_managerial Unmarried White Female 45 United_States rich 41 Private Doctorate 16 Married … Prof_specialty Husband White Male 60 United_States rich : : : : : : : : : : : : : 48,000 records, 16 attributes [Kohavi 1995] Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 14 7
8.
Predicting wealth from
age Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 15 Predicting wealth from age Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 16 8
9.
Wealth from hours
worked Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 17 Wealth from years of education Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 18 9
10.
age, hours →
wealth Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 19 age, hours → wealth Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 20 10
11.
age, hours →
wealth Having 2 inputs instead of one helps in two ways: 1. Combining evidence from two 1d Gaussians 2. Off-diagonal covariance distinguishes class “shape” Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 21 age, hours → wealth Having 2 inputs instead of one helps in two ways: 1. Combining evidence from two 1d Gaussians 2. Off-diagonal covariance distinguishes class “shape” Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 22 11
12.
age, edunum →
wealth Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 23 age, edunum → wealth Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 24 12
13.
hours, edunum →
wealth Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 25 hours, edunum → wealth Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 26 13
14.
Accuracy Copyright © 2001,
Andrew W. Moore Gaussian Bayes Classifiers: Slide 27 An “MPG” example Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 28 14
15.
An “MPG” example
Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 29 An “MPG” example Things to note: •Class Boundaries can be weird shapes (hyperconic sections) •Class regions can be non-simply- connected •But it’s impossible to model arbitrarily weirdly shaped regions •Test your understanding: With one input, must classes be simply connected? Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 30 15
16.
Overfitting dangers • Problem
with “Joint” Bayes classifier: #parameters exponential with #dimensions. This means we just memorize the training data, and can overfit. Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 31 Overfitting dangers • Problem with “Joint” Bayes classifier: #parameters exponential with #dimensions. This means we just memorize the training data, and can overfit. • Problemette with Gaussian Bayes classifier: #parameters quadratic with #dimensions. With 10,000 dimensions and only 1,000 datapoints we could overfit. Question: Any suggested solutions? Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 32 16
17.
General: O(m2)
σ 21 L σ 1m σ 12 σ σ 22 L σ 2m parameters S = 12 M M O M σ L σ 2m σ 2m 1m Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 33 General: O(m2) σ 21 L σ 1m σ 12 σ σ 22 L σ 2m parameters S = 12 M M O M σ L σ 2m σ 2m 1m Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 34 17
18.
σ 21
0 L 0 0 0 Aligned: O(m) 0 σ 22 0 L 0 0 0 0 σ 23 L 0 0 parameters S = M M M M O M L σ 2 m−1 0 0 0 0 0 2 σ m L 0 0 0 Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 35 σ 21 0 L 0 0 0 Aligned: O(m) 0 σ 22 0 L 0 0 0 0 σ 23 L 0 0 parameters S = M M M M O M L σ 2 m−1 0 0 0 0 0 σ 2m L 0 0 0 Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 36 18
19.
σ 2 0
0 L 0 0 Spherical: O(1) 0 σ 0 2 L 0 0 0 0 0 σ2 L cov parameters 0 S = M M M M O M L σ2 0 0 0 0 0 σ 2 L 0 0 0 Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 37 σ 2 0 0 L 0 0 Spherical: O(1) 0 σ2 0 0 L 0 0 0 0 σ2 L cov parameters 0 S = M M M M O M L σ2 0 0 0 0 0 σ 2 L 0 0 0 Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 38 19
20.
BCs that have
both real and categorical inputs? Categorical Real-valued Mixed Real / inputs only inputs only Cat okay Joint BC Gauss BC Predict Inputs Dec Tree Classifier category Naïve BC BC Here??? Joint DE Gauss DE Prob- Inputs Inputs Density ability Estimator Naïve DE Predict Regressor real no. Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 39 BCs that have both real and categorical inputs? Categorical Real-valued Mixed Real / inputs only inputs only Cat okay Joint BC Gauss BC Predict Inputs Dec Tree Classifier category Naïve BC BC Here??? Joint DE Gauss DE Prob- Inputs Inputs Density ability Estimator Naïve DE Easy! Predict Regressor real no. Guess how? Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 40 20
21.
BCs that have
both real and categorical inputs? Categorical Real-valued Mixed Real / inputs only inputs only Cat okay Joint BC Gauss BC Dec Tree Predict Inputs Classifier Gauss/Joint BC category Naïve BC Gauss Naïve BC Joint DE Gauss DE Gauss/Joint DE Prob- Inputs Inputs Density ability Estimator Naïve DE Gauss DE Gauss Naïve DE Predict Regressor real no. Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 41 BCs that have both real and categorical inputs? Categorical Real-valued Mixed Real / inputs only inputs only Cat okay Joint BC Gauss BC Dec Tree Predict Inputs Classifier Gauss/Joint BC category Naïve BC Gauss Naïve BC Joint DE Gauss DE Gauss/Joint DE Prob- Inputs Inputs Density ability Estimator Naïve DE Gauss DE Gauss Naïve DE Predict Regressor real no. Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 42 21
22.
Mixed Categorical /
Real Density Estimation • Write x = (u,v) = (u1 ,u2 ,…uq ,v1 ,v2 … vm-q) Real valued Categorical valued P(x |M)= P(u,v |M) (where M is any Density Estimation Model) Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 43 Joint / Gauss DE sty ich ta Combo re wh … Not su oy? Try our nj e DE to P(u,v |M) = P(u |v ,M) P(v |M) Gaussian with Big “m-q”-dimensional parameters lookup table depending on v Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 44 22
23.
MLE learning of
the Joint / Gauss DE Combo P(u,v |M) = P(u |v ,M) P(v |M) µ v = Mean of u among records matching v Σ v = Cov. of u among records matching v qv = Fraction of records that match v u |v ,M ~ N(µv , Σv ) , P(v |M) = qv Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 45 MLE learning of the Joint / Gauss DE Combo P(u,v |M) = P(u |v ,M) P(v |M) 1 µ v = Mean of u among = ∑ uk records matching v R v k s.t. v k = v Σ v = Cov. of u among =1 ∑ =(u k − µ v )(u k − µ v ) T records matching v R v k s.t.v k v qv = Fraction of records = Rv that match v R R = # records that match v v u |v ,M ~ N(µv , Σv ) , P(v |M) = qv Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 46 23
24.
Gender and Hours
Worked* *As with all the results from the UCI “adult census” dataset, we can’t draw any real-world conclusions since it’s such a non-real-world sample Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 47 Joint / Gauss DE What we just did Combo Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 48 24
25.
Joint / Gauss
BC What we do next Combo Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 49 Joint / Gauss BC Combo p (u , v | M i ) P (Y = i ) P (Y = i | u , v ) = p (u , v ) p (u, | v , M i ) p ( v | M i ) P (Y = i ) = p (u , v ) N (u; µ i , v , S i , v ) q i , v pi = p (u, v ) Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 50 25
26.
Joint / Gauss
BC Combo p (u , v | M i ) P (Y = i ) P (Y = i | u , v ) = p (u , v ) µ i,v = Mean of u among records matching v = p (u, | v , M i ) p ( v | M i ) P (Y = i ) and in which y=i p (u , v ) Σ i,v = Cov. of u among records matching v N (u; µ i , v , S i , v ) q i , v pi and in which y=i = qi,v = Fraction of “y=i” p (u, v ) records that match Rather so-so-notation for v “Gaussian with mean µ i,v and pi = Fraction of records covariance Σ i,v evaluated at u” that match “y=i” Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 51 Gender, Hours→Wealth Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 52 26
27.
Gender, Hours→Wealth Copyright ©
2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 53 Joint / Gauss DE Combo and Joint / Gauss BC Combo: The downside • (Yawn…we’ve done this before…) More than a few categorical attributes blah blah blah massive table blah blah lots of parameters blah blah just memorize training data blah blah blah do worse on future data blah blah need to be more conservative blah Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 54 27
28.
Naïve/Gauss combo for
Density Estimation Categorical Real q m − q ∏ p (u j | M ) ∏ P (v j | M ) p (u , v | M ) = j =1 j =1 u j | M ~ N ( µ j , σ j ) v j | M ~ Multinomia l[q j1 , q j 2 ,..., q jN j ] 2 How many parameters? Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 55 Naïve/Gauss combo for Density Estimation Categorical Real q m − q p (u , v | M ) = ∏ p (u j | M ) ∏ P (v j | M ) j =1 j =1 u j | M ~ N ( µ j , σ j ) v j | M ~ Multinomia l[q j1 , q j 2 ,..., q jN j ] 2 1 ∑ ukj µj = Rk 1 σ 2 = ∑ (u kj − µ j ) 2 j Rk # of records in which v j = h q jh = R Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 56 28
29.
Naïve/Gauss DE Example Copyright
© 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 57 Naïve/Gauss DE Example Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 58 29
30.
Naïve /
p (u , v | Y = i ) P (Y = i ) P (Y = i | u , v ) = Gauss BC p (u , v ) m− q q 1 ∏ p (u j | µ ij , σ ij ) ∏ P (v j | q ij ) P(Y = i ) = 2 p (u , v ) j =1 j =1 m− q q 1 ∏ N (u j ; µ ij ,σ ij ) ∏ qij [ v j ] pi = 2 p (u, v ) j =1 j =1 µ ij = Mean of uj among records in which y=i σ2ij = Var. of uj among records in which y=i qij[h] = Fraction of “y=i” records in which vj = h pi = Fraction of records that match “y=i” Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 59 Gauss / Naïve BC Example Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 60 30
31.
Gauss / Naïve
BC Example Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 61 Learn Wealth from 15 attributes Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 62 31
32.
Learn Wealth from
15 attributes real values discretized Same data, except all to 3 levels Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 63 Learn Race from 15 attributes Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 64 32
33.
What you should
know • A lot of this should have just been a corollary of what you already knew • Turning Gaussian DEs into Gaussian BCs • Mixing Categorical and Real-Valued Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 65 Questions to Ponder • Suppose you wanted to create an example dataset where a BC involving Gaussians crushed decision trees like a bug. What would you do? • Could you combine Decision Trees and Bayes Classifiers? How? (maybe there is more than one possible way) Copyright © 2001, Andrew W. Moore Gaussian Bayes Classifiers: Slide 66 33
Descargar ahora