Computational Learning Theory Lecture Notes for CS Spring Semester

1 Preface This manuscript is a compliation of lecture notes from the graduate level course CS 582, \Computational Learning Theory," I taught a t W ashington University in the spring of 1991. Students taking the course were assumed to have b a c kground in the design and analysis of algorithms as well as good mathematical background. Given that there is no text available on this subject, the course material was drawn from recent research papers. I selected the rst twelve topics and the remainder were selected by the students from a list of provided topics. This list of topics is given at the end of these notes. These notes were mostly written by the students in the class and then reviewed by me. However, there are likely to be errors and omissions, particularly with regard to the references. Readers nding errors in the notes are encouraged to notify me by electronic mail ([email protected]) so that later versions may be corrected. 2 3 Acknowledgements This compilation of notes would not have been possible without the hard work of the following students: I thank Mike Kearns and Umesh Vazirani for providing me with a draft of the scribe notes from their Computational Learning Theory course taught at University of California at Berkeley in the Fall of 1990. Some of my lectures were prepared using their notes. Also most of the homework problems which I g a ve came from the problems used by Ron Rivest for his Machine Learning course at MIT taught during the Falls of 1989 and 1990. I thank Ron for allowing me to include these problems here.