This is an
advanced course in case-based reasoning, with an emphasis on a deeper
understanding of fundamental issues related to the CBR cycle, such as case
representation, case base organization, similarity assessment, case retrieval,
adaptation, and learning. Integrating purely case-based methods with
generalization-based methods is also witin the scope
of the course. The generalization methods may be symbolic or sub-symbolic.
An example of the latter are methods by which CBR enhances sub-symbolic methods
by providing examples of past events for explanatory purposes.
A selected set of book chapters and papers will be discussed.
The concrete focus will to some extent depend on the project and interests of the
students taking the course.
This
doctoral course runs as a self-study course with guidance. We will have two seminar meetings,
in addition to a first introductory meeting. Each seminar meeting will have a
particular topic up for discussion, to be introduced and presented for discussion by the course students.
In the meetings the topic will be covered by a set of papers. The paper presentations should
focus on the motivation for the research described in the corresponding
article, the research goal(s), the research method, the results, and an
evaluation of the results.
In addition, we will organise a set of workshops focusing on
a practical task using the tool myCBR.
The students are assumed to
work with the topics of the
respective papers and tasks in between the meetings, and to browse the Internet or use
other sources to fill in missing details. The students are also encouraged to
arrange group meetings between the scheduled meetings. All students have to
read the papers addressed and prepare for discussions during the course
meetings.
Course schedule:
A start-up meeting will be held on Tuesday February 4th at 10:00 in room 242.
The seminar schedule will be set up in that meeting.
Background assumed,
from earlier courses. Read these (again) before the first meeting!
1.
A.
Aamodt and E. Plaza, 1994: Case-based
reasoning; Foundational issues, methodological
variations, and system approaches. AI Communications, 7(1):39_59.
2.
D.
Aha, 1991: Case-based
learning algorithms. Proceedings of the 1991 DARPA
Case-Based Reasoning Workshop. Morgan Kaufmann.
3.
A.
Aamodt: Knowledge-intensive
case-based reasoning in
Creek. ECCBR
2004. LNAI 3155, Spinger, 2004. pgs.
1-16.
4.
E.L. Rissland: AI and Similarity
AI and Similarity. IEEE Intelligent Systems,
May/June 2006. pp 39-49.
Course material (tentative, likely to be modifid):
Seminar papers:
Similarity
* A. Tversky: Features of Similarity. In: Preference ,
Belief , and Similarity (book), 1st chapter.
* Padraig Cunningham: A Taxonomy of Similarity Mechanisms for Case-Based
Reasoning. University College Dublin, Technical Report
UCD-CSI-2008-01.
* T. Gabel and E. Godehart:
Top-Down Induction of Similarity Measures Using Similarity Clouds.
Seminar papers:
Explainable AI
* F. Sørmo, J. Cassens, A. Aamodt:
Explanation in Case-Based Reasoning–Perspectives and Goals.
* C. Nugent, P. Cunningham:
A Case-Based Explanation System for Black-Box Systems.
* David McSherry:
Explanation in Recommender systems.
* O. Li, H. Liu, C. Chen, C. Rudin:
Deep Learning for Case-Based Reasoning through Prototypes.
NTNU-IDI, January 2020
Kerstin Bach.