Sign in using your ACM Web Account username and password to access premium content if you are an ACM member, Communications subscriber or Digital Library subscriber. Non-members can purchase this article or a copy of the magazine in which it appears.

Children of the 1980s spent hours perfecting the art of playing Namco's Pac-Man computer game. And now computers are teaching each other to play the popular game, which sees a player navigating a maze, trying to gobble up pellets while avoiding four colourful enemies. Researchers have managed to develop a method to allow a computer to give advice and teach skills to another computer in a way that mimics how a real teacher and student might interact - and are demonstrating it with video games. The team, which was led by Matthew Taylor, Washington State University's Allred Distinguished Professor in Artificial Intelligence, used virtual robots called agents to simulate the teacher and student relationship. The student robots initially struggled to learn Pac-Man and a version of the video game StarCraft, but the scientists were able to show that with time the student agent learned the games and surpassed the teacher's abilities.

Many and long were the conversations between Lord Byron and Shelley to which I was a devout and silent listener. During one of these, various philosophical doctrines were discussed, and among others the nature of the principle of life, and whether there was any probability of its ever being discovered and communicated. They talked of the experiments of Dr. Darwin (I speak not of what the doctor really did or said that he did, but, as more to my purpose, of what was then spoken of as having been done by him), who preserved a piece of vermicelli in a glass case till by some extraordinary means it began to move with a voluntary motion. Not thus, after all, would life be given. Perhaps a corpse would be reanimated; galvanism had given token of such things: perhaps the component parts of a creature might be manufactured, brought together, and endued with vital warmth (Butler 1998).

Brandeis has a program in autonomous agents, focusing on multi--agent and multi--robot systems and machine learning, headed by Maja Mataric For details on research directions and a photo of the available robot herd see: http://www.cs.brandeis.edu/dept/faculty/mataric To get more information about the Volen Center for Complex Systems, about the Computer Science Department, and about other faculty, see: http://www.cs.brandeis.edu/dept. For more information about the cognitive science and cognitive neuroscience programs at Brandeis see: http://fechner.ccs.brandeis.edu/cogsci.html The Robotics Institute also offers a Robotics PhD and students from other programs (e.g. Research includes many aspects of mobile robots, computer integrated manufacturing, rapid prototyping, sensors, vision, navigation, learning and architectures.

We have developed a robotic demining system as part of a multi-agent application (AgentStorm) for assisting human commanders in command and control scenarios run in the ModSAF (Modular Semi-Automated Forces) simulation environment. The robotic demining agents cooperatively clear paths, enabling simulated forces to breach minefields. The AgentStorm system is composed of 25 communicating software components developed with the RETSINA agent architecture. It was successfully demonstrated for military observers in November 1999. Within the demining domain, we explore different multi-robot cooperation and communication strategies.

For example, consider a manufacturing scenario in which company X produces tires, but subcontracts the production of lug-nuts to company Y. In order to build a single system to automate (certain aspects of) the production process, the internals of both companies X and Y must be modeled. However, neither company is likely to want to relinquish information and/or control to a system designer representing the other company. Perhaps with just two companies involved, an agreement could be reached, but with several companies involved, MAS is necessary. The only feasible solution is to allow the various companies to create their own agents that accurately represent their goals and interests.

The ontology of a human-like mind: what sorts of states, properties, processes, capabilities can occur in various sorts of minds, e.g. What kinds of architectures can support biological and artificial agents with different kinds of intelligence? This requires a study of design space' and niche space' and their relationships, including the different sorts of trajectories possible in these spaces, e.g. for an individual, for a naturally evolving species, or for a system explicitly modified or repaired by an engineer. To what extent do humans and other agents have simple and uniform architectures, and to what extent do they have hybrid architectures, e.g. Is a human brain an unintelligibly complex morass of mechanisms, or is there sufficient modularity of design to enable us to attain at least a partial understanding of how we work?

Note added 3 Mar 2007: I and some others had some reservations about the document because we felt it did not adequately distinguish degrees that were mainly concerned with training for particular vocations involving use of computers and development of software systems from degrees that were more concerned with teaching the foundations of the subject and preparing future researchers to extend both the foundations and the techniques. This meant, in particular, that some of us felt that there was insufficient emphasis on the mathematical content of computer science and the possibility of doing theoretical (e.g. I personally felt then, and still feel that that expressed a narrowness of vision in the majority of members of the computer science community. The more detailed overview is given in section C, as an indication of the scope of the field. This might be used by reviewers considering degrees in AI, not in order to define what should or should not be in such a degree, but in order to provide some background that might be useful when assessing an AI degree course, or possibly when designing one. There is no implication that everything mentioned here must be included in a degree with "Artificial Intelligence" in its title, or that topics not included here are excluded.

Unlike many so-called'agent toolkits', like PRS/Jack, Mozart, Alice, and several more, that are aimed mainly at development of systems involving large numbers of highly distributed fairly homogeneous relatively'small' agents, SimAgent can be used for such purposes (and was used in that way for a while by Matthias Scheutz at Notre Dame University) but (like ACT-R, COGENT, and the original SOAR) SimAgent is primarily designed to support design and implementation of very complex agents, each composed of very different interacting components (like a human mind) where the whole thing is embedded in an environment that could be a mixture of physical objects and other agents of many sorts, as half-jokingly depicted here: That schema accommodates a wide variety of specific architecture types, which differ according to which mechanisms and information structures occur in which boxes, and how they are connected to one another and to the environment, as described in this overview. The above diagram is misleading in various ways because it suggest that the perception mechanisms and action mechanisms are separate from each other and can only communicate via the'central' mechanisms, whereas it is clear (as James Gibson pointed out in his 1966 book The Senses Considered as Perceptual Systems) action and perception are deeply integrated, e.g. the constant use of saccades, changes of vergence, changes of focus in vision, and the need to move your hand when it is used to perceive shape, texture, weight, flexibility, hardness, etc. of objects. So a more accurate, but less clear depiction of the ideas in the CogAff architecture schema is the following (with thanks to Dean Petters, for help with this diagram, indicating that action and perception mechanisms overlap, as pointed out by J.J.Gibson in 1966(Referenced above). Revised, more realistic CogAff Architecture Schema, e.g. with deeper integration between action and perception The horizontal discs represent (usually "fuzzy" boundaries between different levels of functionality. It is possible for some of the information-processing mechanisms to straddle two or more layers.

NB: Some of the movies use old formats and may not work on your machine without conversion. All are also now available in the WEBM format, which should work, and s houjld be tried first. All the videos are downloadable, under creative commons 3.0 licences. Anyone interested in obtaining and running the toolkit (including some of the demos'frozen' here) should see The Free Poplog Web site. More examples may be added later. Note on Pre-SimAgent demos: Items 10 and 12 below include demos using our pop11 tools before SimAgent was developed. Experience with those projects helped to show the need for the SimAgent tools and the RCLIB graphical interface. If you would like to explore some pre-cursors to SimAgent take a look at the demos near the end of this file.