SlideShare a Scribd company logo
1 of 2
Download to read offline
Statistics 695A: Machine Learning, Fall 2004


Machine Learning                                                    Student Responsibilities
   Machine learning consists of theory, methods, algo-              Seminar Presentation
rithms, models, and software for enabling a computer to
learn from data and to improve learning performance as                Each student should present a journal or conference pa-
the amount of data increases. The area has been developed           per, or a small set of papers on the same topic. Students
by people in many fields such as computer science, statis-           can work in small groups, where the maximum group size
tics, engineering, the biological sciences, and the physical        will be determined by the class size, but each student in
sciences. But applications of machine learning can be car-          the group should carry out a portion of the presentation.
ried out in any field in which it is necessary to learn from         Students are free to select topics as long as they are within
data.                                                               the scope of the course. The choices should be submitted
                                                                    to the TA by Sept. 30, and the papers conveyed as pdf
                                                                    files. The complete schedule of the presentations will be
Prerequisites and Questions                                         announced on October 7. Each presentation will be re-
                                                                    hearsed with the TA a week prior to the class presentation.
  Permission of the instructor is required.                         Each student is expected to read the papers of each of the
  No previous course in machine learning is expected. Pre-          other presentations before it is given. At the end of each
requisites are a basic knowledge of                                 presentation, there will be a discussion session.
     




      probability                                                     Students are encouraged to discuss plans with the TA
     




      mathematics through multi-variable calculus and               and instructor who will be happy to make comments and
      linear algebra                                                suggestions about topics.
     




      least-squares fitting of parametric functions to data
      (Gauss’s machine learning tool)
                                                                    Project Paper
The level of the course will be comparable to that in the
                                                                       Each student should conduct a study, preferably in the
book Machine Learning by Tom M. Mitchell.
                                                                    area of their presentation topic, testing new ideas for tools
  Please send questions about the course or permission to           or using current tools to learn from a set of data. Students
attend to the instructor at wsc@stat.purdue.edu.                    can work in small groups, where the maximum group size
                                                                    will be determined by the class size. Students should
                                                                    report on this work in a short paper of about 8 pages.
Course Orientation and Objectives                                   There should be a discussion of the tasks carried out along
  While the prerequisites do not require previous knowl-            with the interpretation of the results. To write the pa-
edge of machine learning, the course nevertheless has a re-         pers, students should find the most current Web page of a
search orientation. The objectives are to provide students          leading machine learning conference and use the template
with the opportunity                                                and guidelines of the conference to prepare the paper, but
     




      to understand in depth selected areas of machine              keeping it to about 8 pages. The paper should have the
      learning                                                      quality of presentation, at least, of those appearing in such
                                                                    a conference. The paper should be submitted to the TA
     




      to review research in machine learning
                                                                    electronically as pdf by November 30. The papers will be
     




      to experience either the development of machine               reviewed by the TA and comments returned on December
      learning tools or their use to learn from a set of data       7. Students should revise the paper, if necessary, based on
     




      to experience giving a research talk                          these comments, and send to the instructor by December
     




      to experience writing a research paper.                       15.

                                                                1
Students are encouraged to discuss plans with the TA              such tools, and to evaluate the performance of the tool in
and instructor who will be happy to make comments and               learning from the data.
suggestions about topics or available data.                           Robust Learning: What is often overlooked is that in ap-
                                                                    plications a very small fraction of the data can dramati-
                                                                    cally distort the learning output, forcing the results to fol-
Proposed Instructor Lecture Topics                                  low their aberrant behavior. Visualization can often reveal
   N.B. This list could change based on the make-up and             such distortion, but robust learning methods try to prevent
interests of the class.                                             such distortion in the first place.
  Perceptrons and Artificial Neural Nets: Classical tools,             Learning Theory: There will be a sprinkling of “theory”,
one of the first ones that became a part of what we think of         not mathematical derivations, but rather epistemological
today as machine learning. The ANN structure resembles              foundations such as why it helps to think of learning from
biological neural networks, and is made up of multilayer            data as an updating of knowledge, and why it is impor-
networks of perceptrons.                                            tant to learn not just the pattern in a set of data, but the
  Local Learning: This computer intensive approach                  departures of the data from the pattern.
works locally in a multidimensional space, which makes
it amenable to parallel computation. Despite this new-
age usage, basic ideas got started in the 19th and early            Reading about Instructor Lecture Topics
20th Centuries by brilliant actuaries learning about death
                                                                      Various writings on the topics will be available on the
and sickness rates as a function of age. In the 1970s
                                                                    course Web page.
statisticians began what would become a big industry in
statistics research that is often called “nonparametric re-
gression”. Both loess (locally weighted regression) and
projection pursuit regression became widely used tools.             Course Instructor
Machine learning researchers picked up on this work and               William S. Cleveland has been a Professor of Statistics
made many advances in beautiful applications, for exam-             and Computer Science at Purdue University since January
ple, to robotics.                                                   2004. Previous to this he was a Distinguished Member of
  Bayesian Learning: Bayesian learning, single-handedly             Technical Staff in the Statistics and Data Mining Research
intellectually revived by Jimmie Savage in the 1950s and            Department at Bell Labs, Murray Hill.
1960s, is now advancing at a furious pace due to compu-               His areas of research include machine learning, data
tational breakthroughs over the past two decades.                   mining, data visualization, statistical methods and mod-
  Bayesian Networks: Models with a network structure                els, and computer networking.
and computational efficiencies based on certain assump-                Cleveland has introduced tools for local machine learn-
tions. There are a number of marvelous applications to              ing, as well as many visualization tools, that are widely
software systems, for example, developed at Microsoft.              used in engineering, science, medicine, and business. He
An L.A. Times article quotes Bill Gates in an inter-                has participated in the design and implementation of soft-
view: ”Microsoft’s competitive advantage is its expertise           ware for these tools that is now a part of many commercial
in Bayesian networks.”                                              systems. He has been involved in many projects apply-
  Visualization: It cannot be emphasized too strongly how           ing machine learning and visualization tools to data from
important it is to use visualization tools in any application       several fields including environmental science, customer
of learning tools to real data. Visualization helps guide the       opinion polling, visual perception, and computer network-
choice of tools and the estimation of parameters in such            ing.




                                                                2

More Related Content

What's hot

Renuka-Frayer Model for New Generation Science Standards
Renuka-Frayer Model for New Generation Science StandardsRenuka-Frayer Model for New Generation Science Standards
Renuka-Frayer Model for New Generation Science Standardsrekharajaseran
 
S porter article summaries final
S porter article summaries finalS porter article summaries final
S porter article summaries finalsavannahporter1
 
Intc 3610 syllabus spring 2011
Intc 3610 syllabus spring 2011Intc 3610 syllabus spring 2011
Intc 3610 syllabus spring 2011dharvey100
 
Integrating an intelligent tutoring system into a virtual world
Integrating an intelligent tutoring system into a virtual worldIntegrating an intelligent tutoring system into a virtual world
Integrating an intelligent tutoring system into a virtual worldParvati Dev
 
Intelligent tutoring systems (ITS) for online learning
Intelligent tutoring systems (ITS) for online learningIntelligent tutoring systems (ITS) for online learning
Intelligent tutoring systems (ITS) for online learningBrandon Muramatsu
 
A study on the impact of web technologies in teacher education to train the f...
A study on the impact of web technologies in teacher education to train the f...A study on the impact of web technologies in teacher education to train the f...
A study on the impact of web technologies in teacher education to train the f...Dr. C.V. Suresh Babu
 
Strategies and Integrational Pedagogy for Instructional Technology
Strategies and Integrational Pedagogy for Instructional TechnologyStrategies and Integrational Pedagogy for Instructional Technology
Strategies and Integrational Pedagogy for Instructional Technologykendragagnon
 
Issues and Prospects in ICT in Education in Continuing Teacher Professional D...
Issues and Prospects in ICT in Education in Continuing Teacher Professional D...Issues and Prospects in ICT in Education in Continuing Teacher Professional D...
Issues and Prospects in ICT in Education in Continuing Teacher Professional D...rexcris
 

What's hot (9)

Renuka-Frayer Model for New Generation Science Standards
Renuka-Frayer Model for New Generation Science StandardsRenuka-Frayer Model for New Generation Science Standards
Renuka-Frayer Model for New Generation Science Standards
 
S porter article summaries final
S porter article summaries finalS porter article summaries final
S porter article summaries final
 
Intc 3610 syllabus spring 2011
Intc 3610 syllabus spring 2011Intc 3610 syllabus spring 2011
Intc 3610 syllabus spring 2011
 
Integrating an intelligent tutoring system into a virtual world
Integrating an intelligent tutoring system into a virtual worldIntegrating an intelligent tutoring system into a virtual world
Integrating an intelligent tutoring system into a virtual world
 
Btsdsb2018
Btsdsb2018Btsdsb2018
Btsdsb2018
 
Intelligent tutoring systems (ITS) for online learning
Intelligent tutoring systems (ITS) for online learningIntelligent tutoring systems (ITS) for online learning
Intelligent tutoring systems (ITS) for online learning
 
A study on the impact of web technologies in teacher education to train the f...
A study on the impact of web technologies in teacher education to train the f...A study on the impact of web technologies in teacher education to train the f...
A study on the impact of web technologies in teacher education to train the f...
 
Strategies and Integrational Pedagogy for Instructional Technology
Strategies and Integrational Pedagogy for Instructional TechnologyStrategies and Integrational Pedagogy for Instructional Technology
Strategies and Integrational Pedagogy for Instructional Technology
 
Issues and Prospects in ICT in Education in Continuing Teacher Professional D...
Issues and Prospects in ICT in Education in Continuing Teacher Professional D...Issues and Prospects in ICT in Education in Continuing Teacher Professional D...
Issues and Prospects in ICT in Education in Continuing Teacher Professional D...
 

Similar to Statistics 695A: Machine Learning, Fall 2004

Chapter 11 ppt for module 5
Chapter 11 ppt for module 5Chapter 11 ppt for module 5
Chapter 11 ppt for module 5sragasa
 
BE Final Year Project and Seminar Sem VI.pptx
BE Final Year Project and Seminar Sem VI.pptxBE Final Year Project and Seminar Sem VI.pptx
BE Final Year Project and Seminar Sem VI.pptxssuser65a2e8
 
Data Mining and Machine Learning
Data Mining and Machine LearningData Mining and Machine Learning
Data Mining and Machine LearningJakub Ruzicka
 
Project MLExAI: Machine Learning Experiences in AI
Project MLExAI: Machine Learning Experiences in AIProject MLExAI: Machine Learning Experiences in AI
Project MLExAI: Machine Learning Experiences in AIbutest
 
Project MLExAI: Machine Learning Experiences in AI
Project MLExAI: Machine Learning Experiences in AIProject MLExAI: Machine Learning Experiences in AI
Project MLExAI: Machine Learning Experiences in AIbutest
 
Penilaian kendiri-Tugasan 4
Penilaian kendiri-Tugasan 4Penilaian kendiri-Tugasan 4
Penilaian kendiri-Tugasan 4Azhar Yusoff
 
EDUC 4762 Assignment 4.3
EDUC 4762 Assignment 4.3EDUC 4762 Assignment 4.3
EDUC 4762 Assignment 4.3wlavery
 
Collaborative Lesson Plan Farooqi
Collaborative Lesson Plan  FarooqiCollaborative Lesson Plan  Farooqi
Collaborative Lesson Plan FarooqiAysha Farooqi
 
A vavoularis final_presentation
A vavoularis final_presentationA vavoularis final_presentation
A vavoularis final_presentationangelavav
 
Writing Research Paper - Tips For Students
Writing Research Paper - Tips For StudentsWriting Research Paper - Tips For Students
Writing Research Paper - Tips For StudentsRavindra Joshi
 
Tech outline touro_demo - PDF
Tech outline touro_demo - PDFTech outline touro_demo - PDF
Tech outline touro_demo - PDFgibb0
 
3 D Project Based Learning Basics for the New Generation Science Standards
3 D Project Based  Learning Basics for the New Generation Science Standards3 D Project Based  Learning Basics for the New Generation Science Standards
3 D Project Based Learning Basics for the New Generation Science Standardsrekharajaseran
 
Introduction and administrative information (MS Word 97 format)
Introduction and administrative information (MS Word 97 format)Introduction and administrative information (MS Word 97 format)
Introduction and administrative information (MS Word 97 format)butest
 

Similar to Statistics 695A: Machine Learning, Fall 2004 (20)

Chapter 11 ppt for module 5
Chapter 11 ppt for module 5Chapter 11 ppt for module 5
Chapter 11 ppt for module 5
 
BE Final Year Project and Seminar Sem VI.pptx
BE Final Year Project and Seminar Sem VI.pptxBE Final Year Project and Seminar Sem VI.pptx
BE Final Year Project and Seminar Sem VI.pptx
 
Data Mining and Machine Learning
Data Mining and Machine LearningData Mining and Machine Learning
Data Mining and Machine Learning
 
Project MLExAI: Machine Learning Experiences in AI
Project MLExAI: Machine Learning Experiences in AIProject MLExAI: Machine Learning Experiences in AI
Project MLExAI: Machine Learning Experiences in AI
 
Project MLExAI: Machine Learning Experiences in AI
Project MLExAI: Machine Learning Experiences in AIProject MLExAI: Machine Learning Experiences in AI
Project MLExAI: Machine Learning Experiences in AI
 
Media Usability Studies Syllabus
Media Usability Studies Syllabus Media Usability Studies Syllabus
Media Usability Studies Syllabus
 
Penilaian kendiri-Tugasan 4
Penilaian kendiri-Tugasan 4Penilaian kendiri-Tugasan 4
Penilaian kendiri-Tugasan 4
 
EDUC 4762 Assignment 4.3
EDUC 4762 Assignment 4.3EDUC 4762 Assignment 4.3
EDUC 4762 Assignment 4.3
 
libya
libyalibya
libya
 
Collaborative Lesson Plan Farooqi
Collaborative Lesson Plan  FarooqiCollaborative Lesson Plan  Farooqi
Collaborative Lesson Plan Farooqi
 
Data Analysis and Decision Making syllabus
Data Analysis and Decision Making syllabusData Analysis and Decision Making syllabus
Data Analysis and Decision Making syllabus
 
A vavoularis final_presentation
A vavoularis final_presentationA vavoularis final_presentation
A vavoularis final_presentation
 
AIML-MODULE1.pdf
AIML-MODULE1.pdfAIML-MODULE1.pdf
AIML-MODULE1.pdf
 
Writing Research Paper - Tips For Students
Writing Research Paper - Tips For StudentsWriting Research Paper - Tips For Students
Writing Research Paper - Tips For Students
 
Moudle 2
Moudle 2Moudle 2
Moudle 2
 
Cai mpsa software
Cai mpsa softwareCai mpsa software
Cai mpsa software
 
Tech outline touro_demo - PDF
Tech outline touro_demo - PDFTech outline touro_demo - PDF
Tech outline touro_demo - PDF
 
3 D Project Based Learning Basics for the New Generation Science Standards
3 D Project Based  Learning Basics for the New Generation Science Standards3 D Project Based  Learning Basics for the New Generation Science Standards
3 D Project Based Learning Basics for the New Generation Science Standards
 
Introduction and administrative information (MS Word 97 format)
Introduction and administrative information (MS Word 97 format)Introduction and administrative information (MS Word 97 format)
Introduction and administrative information (MS Word 97 format)
 
Information Skills
Information SkillsInformation Skills
Information Skills
 

More from butest

EL MODELO DE NEGOCIO DE YOUTUBE
EL MODELO DE NEGOCIO DE YOUTUBEEL MODELO DE NEGOCIO DE YOUTUBE
EL MODELO DE NEGOCIO DE YOUTUBEbutest
 
1. MPEG I.B.P frame之不同
1. MPEG I.B.P frame之不同1. MPEG I.B.P frame之不同
1. MPEG I.B.P frame之不同butest
 
LESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALLESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALbutest
 
Timeline: The Life of Michael Jackson
Timeline: The Life of Michael JacksonTimeline: The Life of Michael Jackson
Timeline: The Life of Michael Jacksonbutest
 
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...butest
 
LESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALLESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALbutest
 
Com 380, Summer II
Com 380, Summer IICom 380, Summer II
Com 380, Summer IIbutest
 
The MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
The MYnstrel Free Press Volume 2: Economic Struggles, Meet JazzThe MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
The MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazzbutest
 
MICHAEL JACKSON.doc
MICHAEL JACKSON.docMICHAEL JACKSON.doc
MICHAEL JACKSON.docbutest
 
Social Networks: Twitter Facebook SL - Slide 1
Social Networks: Twitter Facebook SL - Slide 1Social Networks: Twitter Facebook SL - Slide 1
Social Networks: Twitter Facebook SL - Slide 1butest
 
Facebook
Facebook Facebook
Facebook butest
 
Executive Summary Hare Chevrolet is a General Motors dealership ...
Executive Summary Hare Chevrolet is a General Motors dealership ...Executive Summary Hare Chevrolet is a General Motors dealership ...
Executive Summary Hare Chevrolet is a General Motors dealership ...butest
 
Welcome to the Dougherty County Public Library's Facebook and ...
Welcome to the Dougherty County Public Library's Facebook and ...Welcome to the Dougherty County Public Library's Facebook and ...
Welcome to the Dougherty County Public Library's Facebook and ...butest
 
NEWS ANNOUNCEMENT
NEWS ANNOUNCEMENTNEWS ANNOUNCEMENT
NEWS ANNOUNCEMENTbutest
 
C-2100 Ultra Zoom.doc
C-2100 Ultra Zoom.docC-2100 Ultra Zoom.doc
C-2100 Ultra Zoom.docbutest
 
MAC Printing on ITS Printers.doc.doc
MAC Printing on ITS Printers.doc.docMAC Printing on ITS Printers.doc.doc
MAC Printing on ITS Printers.doc.docbutest
 
Mac OS X Guide.doc
Mac OS X Guide.docMac OS X Guide.doc
Mac OS X Guide.docbutest
 
WEB DESIGN!
WEB DESIGN!WEB DESIGN!
WEB DESIGN!butest
 

More from butest (20)

EL MODELO DE NEGOCIO DE YOUTUBE
EL MODELO DE NEGOCIO DE YOUTUBEEL MODELO DE NEGOCIO DE YOUTUBE
EL MODELO DE NEGOCIO DE YOUTUBE
 
1. MPEG I.B.P frame之不同
1. MPEG I.B.P frame之不同1. MPEG I.B.P frame之不同
1. MPEG I.B.P frame之不同
 
LESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALLESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIAL
 
Timeline: The Life of Michael Jackson
Timeline: The Life of Michael JacksonTimeline: The Life of Michael Jackson
Timeline: The Life of Michael Jackson
 
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
 
LESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALLESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIAL
 
Com 380, Summer II
Com 380, Summer IICom 380, Summer II
Com 380, Summer II
 
PPT
PPTPPT
PPT
 
The MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
The MYnstrel Free Press Volume 2: Economic Struggles, Meet JazzThe MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
The MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
 
MICHAEL JACKSON.doc
MICHAEL JACKSON.docMICHAEL JACKSON.doc
MICHAEL JACKSON.doc
 
Social Networks: Twitter Facebook SL - Slide 1
Social Networks: Twitter Facebook SL - Slide 1Social Networks: Twitter Facebook SL - Slide 1
Social Networks: Twitter Facebook SL - Slide 1
 
Facebook
Facebook Facebook
Facebook
 
Executive Summary Hare Chevrolet is a General Motors dealership ...
Executive Summary Hare Chevrolet is a General Motors dealership ...Executive Summary Hare Chevrolet is a General Motors dealership ...
Executive Summary Hare Chevrolet is a General Motors dealership ...
 
Welcome to the Dougherty County Public Library's Facebook and ...
Welcome to the Dougherty County Public Library's Facebook and ...Welcome to the Dougherty County Public Library's Facebook and ...
Welcome to the Dougherty County Public Library's Facebook and ...
 
NEWS ANNOUNCEMENT
NEWS ANNOUNCEMENTNEWS ANNOUNCEMENT
NEWS ANNOUNCEMENT
 
C-2100 Ultra Zoom.doc
C-2100 Ultra Zoom.docC-2100 Ultra Zoom.doc
C-2100 Ultra Zoom.doc
 
MAC Printing on ITS Printers.doc.doc
MAC Printing on ITS Printers.doc.docMAC Printing on ITS Printers.doc.doc
MAC Printing on ITS Printers.doc.doc
 
Mac OS X Guide.doc
Mac OS X Guide.docMac OS X Guide.doc
Mac OS X Guide.doc
 
hier
hierhier
hier
 
WEB DESIGN!
WEB DESIGN!WEB DESIGN!
WEB DESIGN!
 

Statistics 695A: Machine Learning, Fall 2004

  • 1. Statistics 695A: Machine Learning, Fall 2004 Machine Learning Student Responsibilities Machine learning consists of theory, methods, algo- Seminar Presentation rithms, models, and software for enabling a computer to learn from data and to improve learning performance as Each student should present a journal or conference pa- the amount of data increases. The area has been developed per, or a small set of papers on the same topic. Students by people in many fields such as computer science, statis- can work in small groups, where the maximum group size tics, engineering, the biological sciences, and the physical will be determined by the class size, but each student in sciences. But applications of machine learning can be car- the group should carry out a portion of the presentation. ried out in any field in which it is necessary to learn from Students are free to select topics as long as they are within data. the scope of the course. The choices should be submitted to the TA by Sept. 30, and the papers conveyed as pdf files. The complete schedule of the presentations will be Prerequisites and Questions announced on October 7. Each presentation will be re- hearsed with the TA a week prior to the class presentation. Permission of the instructor is required. Each student is expected to read the papers of each of the No previous course in machine learning is expected. Pre- other presentations before it is given. At the end of each requisites are a basic knowledge of presentation, there will be a discussion session.   probability Students are encouraged to discuss plans with the TA   mathematics through multi-variable calculus and and instructor who will be happy to make comments and linear algebra suggestions about topics.   least-squares fitting of parametric functions to data (Gauss’s machine learning tool) Project Paper The level of the course will be comparable to that in the Each student should conduct a study, preferably in the book Machine Learning by Tom M. Mitchell. area of their presentation topic, testing new ideas for tools Please send questions about the course or permission to or using current tools to learn from a set of data. Students attend to the instructor at wsc@stat.purdue.edu. can work in small groups, where the maximum group size will be determined by the class size. Students should report on this work in a short paper of about 8 pages. Course Orientation and Objectives There should be a discussion of the tasks carried out along While the prerequisites do not require previous knowl- with the interpretation of the results. To write the pa- edge of machine learning, the course nevertheless has a re- pers, students should find the most current Web page of a search orientation. The objectives are to provide students leading machine learning conference and use the template with the opportunity and guidelines of the conference to prepare the paper, but   to understand in depth selected areas of machine keeping it to about 8 pages. The paper should have the learning quality of presentation, at least, of those appearing in such a conference. The paper should be submitted to the TA   to review research in machine learning electronically as pdf by November 30. The papers will be   to experience either the development of machine reviewed by the TA and comments returned on December learning tools or their use to learn from a set of data 7. Students should revise the paper, if necessary, based on   to experience giving a research talk these comments, and send to the instructor by December   to experience writing a research paper. 15. 1
  • 2. Students are encouraged to discuss plans with the TA such tools, and to evaluate the performance of the tool in and instructor who will be happy to make comments and learning from the data. suggestions about topics or available data. Robust Learning: What is often overlooked is that in ap- plications a very small fraction of the data can dramati- cally distort the learning output, forcing the results to fol- Proposed Instructor Lecture Topics low their aberrant behavior. Visualization can often reveal N.B. This list could change based on the make-up and such distortion, but robust learning methods try to prevent interests of the class. such distortion in the first place. Perceptrons and Artificial Neural Nets: Classical tools, Learning Theory: There will be a sprinkling of “theory”, one of the first ones that became a part of what we think of not mathematical derivations, but rather epistemological today as machine learning. The ANN structure resembles foundations such as why it helps to think of learning from biological neural networks, and is made up of multilayer data as an updating of knowledge, and why it is impor- networks of perceptrons. tant to learn not just the pattern in a set of data, but the Local Learning: This computer intensive approach departures of the data from the pattern. works locally in a multidimensional space, which makes it amenable to parallel computation. Despite this new- age usage, basic ideas got started in the 19th and early Reading about Instructor Lecture Topics 20th Centuries by brilliant actuaries learning about death Various writings on the topics will be available on the and sickness rates as a function of age. In the 1970s course Web page. statisticians began what would become a big industry in statistics research that is often called “nonparametric re- gression”. Both loess (locally weighted regression) and projection pursuit regression became widely used tools. Course Instructor Machine learning researchers picked up on this work and William S. Cleveland has been a Professor of Statistics made many advances in beautiful applications, for exam- and Computer Science at Purdue University since January ple, to robotics. 2004. Previous to this he was a Distinguished Member of Bayesian Learning: Bayesian learning, single-handedly Technical Staff in the Statistics and Data Mining Research intellectually revived by Jimmie Savage in the 1950s and Department at Bell Labs, Murray Hill. 1960s, is now advancing at a furious pace due to compu- His areas of research include machine learning, data tational breakthroughs over the past two decades. mining, data visualization, statistical methods and mod- Bayesian Networks: Models with a network structure els, and computer networking. and computational efficiencies based on certain assump- Cleveland has introduced tools for local machine learn- tions. There are a number of marvelous applications to ing, as well as many visualization tools, that are widely software systems, for example, developed at Microsoft. used in engineering, science, medicine, and business. He An L.A. Times article quotes Bill Gates in an inter- has participated in the design and implementation of soft- view: ”Microsoft’s competitive advantage is its expertise ware for these tools that is now a part of many commercial in Bayesian networks.” systems. He has been involved in many projects apply- Visualization: It cannot be emphasized too strongly how ing machine learning and visualization tools to data from important it is to use visualization tools in any application several fields including environmental science, customer of learning tools to real data. Visualization helps guide the opinion polling, visual perception, and computer network- choice of tools and the estimation of parameters in such ing. 2