CMSC 691 M Agent Architectures Multi Agent Systems

CMSC 691 M Agent Architectures & Multi. Agent Systems UMBC Prof. Marie des. Jardins Spring 2002

Course information n Prof des. Jardins q n TA – Gunjan Kalra q n ECS 216, x 53967, mariedj@cs. umbc. edu gkalra 1@cs. umbc. edu Class mailing list q q cs 691 m@listproc. umbc. edu To subscribe, send email to listproc@listproc. umbc. edu with the line: n subscribe cs 691 m Your Name

Today’s overview n n Class structure and policies What’s an agent? Agent exercise Next class

Class structure: Syllabus n n Course page: cmsc 691 m. html Course syllabus: schedule. html

Class structure: Participation n This is a discussion class q q n Reading must be done in advance Participation counts—a lot 45% of grade is related to class participation q q Reading summaries (15% plus bonus points) Class participation (20%) Discussion leaders (5%) Note takers (5%)

Class structure: Agent architecture project n Midterm paper/project: 25% of grade q q q q Compare two architectures Investigate in more depth than in class Can download software, do extra reading, try implementing part of the architecture, … Proposal due Feb. 21 (5% of paper) Draft due Mar. 14 (40% of paper) Review due Apr. 2 (5% of class) Final draft due Apr. 11 (55% of paper)

Class structure: MAS project n n n Agent to participate in multi-agent environment Most likely domains: TAC or Robo. Cup Domain presentation will be given on March 14 Dry run opportunity on May 7 Tournament and papers/presentations at time final exam is scheduled

Policies n n Grading and academic honesty: grading. ps Plagiarism, citations

What’s an agent? n Weiss, p. 29 [after Wooldridge and Jennings]: q n Russell and Norvig, p. 7: q n “An agent is a computer system that is situated in some environment, and that is capable of autonomous action in this environment in order to meet its design objectives. ” “An agent is just something that perceives and acts. ” Rosenschein and Zlotkin, p. 4: q “The more complex the considerations that [a] machine takes into account, the more justified we are in considering our computer an ‘agent, ’ who acts as our surrogate in an automated encounter. ”

What’s an agent? II Ferber, p. 9: n “An agent is a physical or virtual entity q a) b) c) d) e) f) g) h) i) Which is capable of acting in an environment, Which can communicate directly with other agents, Which is driven by a set of tendencies…, Which possesses resources of its own, Which is capable of perceiving its environment…, Which has only a partial representation of this environment…, Which possesses skills and can offer services, Which may be able to reproduce itself, Whose behavior tends towards satisfying its objectives, taking account of the resources and skills available to it and depending on its perception, its representations and the communications it receives. ”

OK, so what’s an environment? n Isn’t any system that has inputs and outputs situated in an environment of sorts?

What’s autonomy, anyway? n Jennings and Wooldridge, p. 4: q n “[In contrast with objects, we] think of agents as encapsulating behavior, in addition to state. An object does not encapsulate behavior: it has no control over the execution of methods – if an object x invokes a method m on an object y, then y has no control over whether m is executed or not – it just is. In this sense, object y is not autonomous, as it has no control over its own actions…. Because of this distinction, we do not think of agents as invoking methods (actions) on agents – rather, we tend to think of them requesting actions to be performed. The decision about whether to act upon the request lies with the recipient. ” Is an if-then-else statement sufficient to create autonomy?

So now what? n If those definitions aren’t useful, is there a useful definition? Should we bother trying to create “agents” at all?

Agent exercise n Pick a card, any card…

After-action review or post-mortem, as the case may be… n n n Did the class (agent community) find a consistent solution? How many agents had an instantiation? How many constraints were violated? q n Why those ones? Any theories? What’s hard about this problem?

Next class n n n Reading: Weiss Prologue and Chapter 1 NO reading summary this time, but you should come with some additional questions of your own Questions for Day 2: q q Characterize today’s exercise in terms of the agent characteristics on page 4 When is something an agent, and when is it just a piece of software? Is there any difference? Is it worth having “agents” that aren’t “intelligent agents”? What do you want to get out of this class? What part of the syllabus are you most excited about? Least excited?