Overview Advanced AI

1
Overview Advanced AI
History of AI thr. 1995
AI and the Web Traditional
Clustering
Graph Mining Spatial and Spatio-Temporal Data
Mining Reinforcement Learning and Learning to
Lean Shape-based Image Retrieval
Hyla-Tree Frog
2
General Thoughts and Teaching Philosophy I

• Focus of the course is providing more indepth
  knowledge in the areas mentioned on the previous
  slight and to learn how to read, summarize,
  present, and evaluate scientific papers.
• Interactive discussion of papers and research
  topics
• No cheating! No cheating! No cheating!
• Teaching Philosophy
• You will have to face some criticism otherwise,
  you will not learn anything. Learning without
  exposing yourself to errors is impossible!!
• If you do not know what you do wrong, it is hard
  to improve!
• No matter if you like it are not, you will have
  to talk a lot in this course.
• During the course you will make a lot of
  informal, unstructured presentations.

You do Something
Learn
Feedback
3
General Thoughts and Teaching Philosophy II

• No projects and only 2 quizzes first week of
  March and 3rd week of April.
• Learning by doing!!
• I am aware that most of you are not too
  experienced in these matters consequently, my
  expectations are initially quite low.
• Not reading papers that will be discussed on a
  particular day is not acceptable!
• One objective if this course is to describe what
  other do in your own words consequently, no
  copying from any sources (web or class mates).
• If you face particular or unusual problems when
  taking this course talk to me during my office
  hours or send me an e-mail.
• During the course you will also make 2 more
  formal presentations.

4
General Thoughts and Teaching Philosophy III

• You will also get some exposure concerning
  writing abstracts, summaries, introductions,
  white paper, and conclusions.
• Learning to write included to know what people
  expect concerning what you write and how what you
  write will be judged.
• In this course, we will try it several teaching
  strategies, some of which will be revised or even
  abandoned as the course progresses.
• Is a by product you will hopefully get a better
  understanding on how to conduct a scientific
  project and on how to summarize and present its
  results.
• Occasionally, Dr. Eick will present 10-40 minute
  lectures that provide background knowledge
  concerning the papers that will be discussed
  next.

5
Topics Covered

• History of AI thr. 1995 the 1994 Turing Award
  Lectures
• AI and the Web how to assess the relevance of
  webpages?
• Traditional Clustering
• Shape-based Image Retrieval
• Spatial and Spatio-Temporal Data Mining
• Reinforcement Learning and Learning to Lean
• Graph Mining
• Papers allocation to areas I 2 II 2-4
  III3-4 IV1 V3-5 VI2-3 VII0-1.

6
Paper Reading List COSC 7363

• AI through 1995 Edward Feigenbaums and Raj
  Reddys 1994 Turing Award Lectures in CACM, May
  1996, pages 97-112. Likely, only Feigenbaums
  lecture How the What becomes the How will
  be discussed.
• Page Brin (Google Publication), The PageRank
  Citation Ranking Bringing Order to the Web,
  1998 walkthrough paper perhaps this is a better
  source of the work Page Brin (Google
  Publication), The Anatomy of a Large-Scale
  Hypertextual Web Search Engine,1999.
• CV of Sergey Brin potcast of 2005 Interview
  with Sergey Brin (http//www.itconversations.com/s
  hows/detail795.html)
• Hanghang Tong, Christos Faloutsos, and Jia-Yu
  Pan, Fast Random Walk with Restart and Its
  Applications, Proc. ICDM Conference, Hong Kong,
  China, Dec. 2006 best research paper award 2
  student- supervised walkthrough
• Langville Meyer, Deeper Inside Page Rank,
  Internet Mathematics, Vol. 1, No. 3, 335-380,
  2004 potential student presentation paper
• Original DBSCAN Paper 2-student-supervised
  walkthough
• Rousseaux Original Silhouette Paper walkthrough
  learn how to write an introduction and an
  abstract
• Likely, Original EM Paper McLachlan, G. and
  Peel, D. (2000). Finite Mixture Models. J. Wiley,
  New York. teams of 2 read the paper, learn how to
  write a conclusion
• Clustering with Bregman Divergences by A.
  Banerjee, S. Merugu, I. S. Dhillon, and J. Ghosh,
  in Journal of Machine Learning Research, vol. 6,
  pages 1705-1749, October 2005

7
Paper Reading List COSC 7363

• Cyrus Shahabi, Maytham Safar, An experimental
  study of alternative shape-based image retrieval
  techniques, Multimedia Tools and Applications,
  Springer Netherlands, November 2006.
• S. Shekhar, P. Zhang, Y. Huang, R. Vatsavai,
  Trends in Spatial Data Mining, book chapter in
  Data Mining Next Generation Challenges and
  Future Directions, H. Kargupta, A. Joshi, K.
  Sivakumar, and Y. Yesha(eds.), AAAI/MIT Press,
  2003.
• Some Spatial Statistics Paper
• Co-location Mining with Rare Spatial Features by
  Yan Huang, Jian Pei, and Hui Xiong published in
  Journal of GeoInformatica, vol. 10, issue 3,
  2006.
• Mirco Nanni, Dino Pedreschi. Time-focused
  density-based clustering of trajectories of
  moving objects. in Journal of Intelligent
  Information Systems (JIIS), 27(3)267-289,
  Special Issue on "Mining Spatio-Temporal Data",
  2006.
• H. Cao, N. Mamoulis, and D. W. Cheung, 
  "Discovery of Periodic Patterns in Spatiotemporal
  Sequences," IEEE Transactions on Knowledge and
  Data Engineering (TKDE), to appear.
• Reinforcement Learning A Survey by Kaelbling, L.
  P. and Littman, M. L. in JAIR, 1996
  (http//www.cs.cmu.edu/afs/cs/project/jair/pub/vol
  ume4/kaelbling96a-html/rl-survey.html )
• Very Likely Something from the Thrun book
  Learning to Learn, 1998.
• If enough time left Yan, X. and Han, J. 2002.
  gSpan Graph-Based Substructure Pattern Mining,
  in Proceedings of the 2002 IEEE International
  Conference on Data Mining (ICDM '02) Washington,
  DC.

8
Teaching Plan Next 5 Weeks

• Janurary 16 Course Overview (see also
  http//www2.cs.uh.edu/ceick/7363/7363.html)
• January 18 (Read) Feigenbaum Paper ? How to read
  a paper
• January 23 Prepare Slow PaperWalkThrough
  Page/Brin Paper
• January 25 Mostly Lecture
• January 30 Prepare Slow PaperWalk Through ICDM06
  Best Research Paper Award
• Feb. 1 TBDL --- Brin/Page Background?!?
• Feb. 6 Likely Reddy Turing Award Paper
• Feb. 8 Student Guided Discussion DBSCAN paper
• Feb. 13 Student Guided Discussion Silhouette
  Paper
• Feb. 15 Student Presentation Langville/Mayer
  Paper (optional)
• Feb. 20 On Writing Abstracts, Introductions,
  Conclusions
• Feb. 22 Bregman Divergences Paper

9
Course Activities

• A lot of informal presentations and discussions
• 1 formal presentation about a paper covered in
  the course
• Writing abstracts, introductions, conclusions and
  paper reviews --- learning by doing
• 2 Quizes that ask questions about papers we have
  read
• Discussions
• Learning how to read, summarize, present, and
  review papers.
• Background knowledge on how to perform a
  research project and on how to be successful in
  your research / career.
• Discussing many other, entertaining things---such
  as the Giant Squid, life of Sergey Brin--- most
  of which are still related to one of the above
  activities.

10
Forms of Covering Papers in the Course

• Papers will be discussed, presented in many
  different forms in this course
• Slow Walk Through (I only plan to have 3 of
  those!!)
• Guided Walk Through
• Other Walk Throughs (I did not consider yet!)
• By answering a given set of quesitons.
• Just Discussion
• 1-Page (5-page) Summary of a Paper
• Professional Powerpoint Presentation
• Profession Paper Review (? April 2007)

11
Slow Walk Throughs

• Used for the first two AI and the web papers
• Paper will be discussed paragraph by paragraph
• Very slow!! Therefore, there will be only 2-4 of
  those
• Course participants are responsible for sections
  of the paper. Responsibilities include
• Lead discussion
• Present short summaries for boring sections to
  speed up things
• Ask questions about things they do not understand
• Prepare review questions for the other students
  that will be discussed either immediately or
  after a delay.
• Everybody should read the paper carefully
  including the sections you are not responsible
  for. It might be a good idea to create brief
  summaries for the read sections and to capture,
  what you do not understand, in form of questions.
  
• If you finished reading the paper try to come up
  with your own evaluation of the paper
• We will not only discuss the contents of the
  paper, but also address the question why an
  author writes a paper in a particular way and
  how the presentation of the discussed paper
  could be improved / made more convincing.
• Additionally, issues on how to write a paper will
  be discussed during slow walk throughs --- these
  matters are Dr. Eicks responsibility..

12
Sergey Brin CV (?????? ?????????? ????) (see
also http//en.wikipedia.org/wiki/Sergey_Brin )

• Sergey Brin is co-founder and President,
  Technology at Google. Originally a native of
  Moscow, he received a bachelor of science degree
  with honors in mathematics and computer science
  from the University of Maryland at College Park.
  He is currently on leave from the Ph.D. program
  in computer science at Stanford University, where
  he received his master's degree. Sergey is a
  recipient of a National Science Foundation
  Graduate Fellowship as well as an honorary MBA
  from Instituto de Empresa. It was at Stanford
  that he met Larry Page and worked on the project
  that became Google. Together they founded Google
  Inc. in 1998, and Sergey continues to share
  responsibility for day-to-day operations with
  Larry Page and Eric Schmidt.
• Sergey's research interests include search
  engines, information extraction from unstructured
  sources, and data mining of large text
  collections and scientific data. He has published
  more than a dozen academic papers and has been a
  featured speaker at several international
  academic, business and technology forums,
  including the World Economic Forum and the
  Technology, Entertainment and Design Conference.
  He has shared his views on the technology
  industry and the future of search on the Charlie
  Rose Show, CNBC, and CNNfn. In 2004, he and Larry
  Page were named "Persons of the Week" by ABC
  World News Tonight.
• See also http//money.cnn.com/magazines/fortune/fo
  rtune_archive/2006/10/02/8387489/index.htm?postver
  sion2006100210 for more about Googles work
  philosophy.

Sergey
13
Assignments Tong Slow Walk Through

• .Remark scheduled for Tu., January 30, 2007

14
On Exploring Boundaries

• There is always a limit on what you can still do
  / cannot do.
• However, these boundaries are dynamic and can be
  changed through training
• Analogy Training for an Olympic Marathon
• Several of the papers we will read will be not
  easy to understand
• If you read the papers we discuss in the course
  now you will understand 25 of the contents
  hopefully, on May 1, 2007 you will understand 70
  of the papers discussed
• You should believe into yourself that you can
  extend and challenge these boundaries. A
  statement like I will never understand this
  paper is not practical is not productive
• Being not afraid of boundaries is particularly
  important for personal growth and research.

15
Analogy Training to win a Medal at an Olympic
Marathon

• You need to have some basic talent to have a
  chance --- but a lot of people have talent
• You have to be committed and have to believe that
  you have a chance
• You have to have some luck
• At least 50 depends on the training you do.
  Dilemma
• If you train too hard you get injured
• If you train a little you have no chance
• Marathon training is about extending your
  boundaries without getting hurt...
• To be successful you need a coach (or even a team
  of coaches)

What you can do in a year
What you can do today
16
Some Ingredients for Success
Time, Money, Food, Friends, Place to live,
Talent
Training
Self Confidence and Commitment
Persistance
Luck
17
Another Example Finding the Giant Squid
10 meters
18 meters
Kubodera said catching the squid on film was the
result of 10 years of sleuthing.
We knew that sperm whales fed on the squid, and
we knew when and how deep they dived, Kubodera
said. So we used them to lead us to the squid.
Giant Squid Sperm Whale
Story http//www.msnbc.msn.com/id/9503272/ On
sperm whales http//www.oceanicresearch.org/sperm
whales.htm Another Sperm Whale Photo
http//nmml.afsc.noaa.gov/gallery/cetaceans/pm-13_
sperm.htm