In this report, I focused on my presentation for the upcoming ICAI ‘09 conference in Las Vegas.  It has currently 18 slides.

I am also viewing the video lectures from the Summer Schools in Logic and Learning from Video Lectures dot Net.  Thus, I have completed the Introduction to Logic video lectures.

I was also reading Michalski’s article on A Theory and Methodology of Induction Learning. Basically it was an article describing the induction process and some the issues associated with induction such as descriptive language, background information, examples, generalization, and hypothesis space.  All this sounded familar when I was reading the two books on ILP.  Induction is a common theme with the two books.  Various familar algorithms are discussed amongst this material.  It appears there are no easy answers in the induction of rules from examples.

I completed reading Jaime Carbonell article Learning By Analogy: Formulating and Generalizing Plans from Past Experience.  In summary, the article discusses using analogous situations to solve problems and develop plans based on analogy or similarity.  This is true since I have developed many patterns in solving various problems in my life.  For example, when learned about cross pinning in chess, I was able to find a position in the game in which I applied the cross pin.  This was a form of transfer learning or learning by analogy.  The same principle can be applied to machine learning according to Carbonell.  He discusses a theory, a formulation, an example of this process in machine learning, and some success in this arena.

I continued research into the Advice Language.  Although interesting, the advice language has not been widely utilized as the researchers had hoped.  Various studies by Michie, Bratko, Niblett, and Kopec had realized some interesting results, but not enough by the AI community.  Research into chess endgames yielded some interesting results – for example, the king and pawn versus king (KPK) can be won in 19 moves, and the king and rook versus the king (KRK) can be in 16 moves.  Research with the king and rook versus the king and knight (KRKN) was performed by Michie, Bratko, Kopec and Niblett using the advice language.  In addition, Quinlan performed classification research with same endgame to determine a decision if it could be determined that a KRKN could won in three moves or less.

The important conferences this month was the joint ICML, UAI, and COLT conferences.  The tutorials was interesting as well as some of the workshop topics.  The ICML was from 6/14 to 6/18, with the joint workshop on 6/19.  The UAI and COLT conferences was from 6/18 to 6/21.

For the fourth month in a row, my blog has reached another monthly high with 310. hits  The hotest blog entries did not change from last month’s report.  Wumpus World and Wumpus World Revisited are most active blogs, accounting for 20 percent of this year hits.

Thank you for support.