Department of Computer Science
IT 803 Instructable Agents
· Overview of the course’s objective and requirements:
o Hands-on experience with designing and developing an end-to-end instructable agent
o Writing of a research paper focusing on one aspect of the agent development process
· Overview of the agent development process and of the Disciple shell
· Introduction of the agent to be developed in this course: An assistant for choosing a PhD thesis advisor.
Assignment 1: Literature search on how to choose a PhD thesis advisor (due February 4th)
· Student presentations of the results of the literature search (the studied papers and associated presentations will be part of the final report of the course)
· Specification of the intelligent assistant to be developed
· Outline of the main stages and deadlines of the agent development process
Assignment 2: Preliminary modeling of the process of choosing a PhD thesis advisor, based on the task reduction paradigm (due February 11th)
· Specification of an example of a PhD thesis advisor
· Development of the top level modeling for choosing a PhD Thesis advisor
· Partitioning of the domain and assignment of the partitions to individual students (each student will develop an agent corresponding to the assigned partition; all these agents will share a common ontology and will finally be integrated into a more powerful agent)
· Discussion of individual research papers
Assignment 3: First draft of the individual research paper (due February 18th)
· Student presentations of the first drafts of their individual research papers
Assignment 4: Development of the modeling for the student assigned partition based on the specified example of a PhD thesis advisor (due February 25th)
· Discussion and integration of the problem solving models developed by students
· Specification of an additional example of a PhD thesis advisor
Assignment 5: Refined example, refined modeling to include the additional example, and ontology specification for the entire modeling (due March 4th)
· Discussion of the additional example, refined modeling and ontology specification
Assignment 6: Ontology development by individual students (due March 18th)
· Discussion of the developed ontologies and their integration into a common ontology
Assignment 7: Revision of individual ontology fragments, integration of the ontology fragments and scripts development, including report generation (due March 25th)
· Discussion of the integrated ontology and of the developed scripts
Assignment 8: Mid-term version of the individual research paper (due April 1st)
· Student presentations of the mid-term versions of the individual research papers.
Assignment 9: Elicitation and modeling for the first example of PhD thesis advisor (due April 8th)
· Discussion of the elicited example and modeling
Assignment 10: Task formalization, rule learning and problem solving for the first example of PhD thesis advisor (due April 12th)
April 15 to be rescheduled for Friday April 12, if possible
· Discussion and demo of learning and problem solving.
Assignment 11: Elicitation and training for the second example of a PhD thesis advisor (due April 19th).
April 22 to be rescheduled for Friday April 19, if possible
· Presentation of the developed agents and discussion of their future developments
· Development of the final agent, integrating the capabilities of the developed agents.
· Discussion of the strengths and weaknesses of the agent development process
· Lessons learned
· Future developments.
Assignment 12: Final version of individual research paper (due May 13th)
· Research papers presentations