Chapman, D., and Agre, P. E., 1987. Abstract reasoning as emergent from concrete activity, Reasoning about Actions and Plans (M. P. Georgeff and A. L. Lansky, eds.), Los Altos, CA: Morgan Kaufmann, 411-424.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....and the PENGI system [Agre and Chapman, 1987] in particular. The REX system is specifically designed to be a robot programming language, and as such is defended by its ability to fiexibly describe very general robotic tasks. The PENGI system is an implementation of a theory of situated activity , [Chapman and Agre, 1986] and is defended, not by its contribution to robot programming, but by the theory s power to describe real world behavior. 1 The RAP system is both a programming language and a proposal for structuring behavior descriptions and can be defended on both counts. The RAP representation syntax is ....

....that each behavior maintains only as much state as required to carry out its local, hardwired goals. There is no overall control structure, no explicit plan, and no internal model of the world [Brooks, 1987] Chapman and Agre argue for a similar model of robot control they call situated activity [Chapman and Agre, 1986]. The implementation of their model in the PENGI system does not have an explicit layered structure, but rather consists of a visual routine processor and a decision network built up of simple logic gates [Agre and Chapman, 1987] Information about the world comes in through the visual processor ....

David Chapman and Philip E. Agre. Abstract reasoning as emergent from concrete activity. In Workshop o Plaig ad Rcasoig about Actions, Portland, Oregon, June 1986.

....abstract resources. Firby [Firby 87] proposed a reactive action package (RAP) that pursues a planning goal until that goal has been achieved. Failures caused by unexpected world situations are relegated to the underlying hardware interface in an attempt to remodel the world. Chapman and Agre [Chapman 86] do reactive planning organised around situation action like rules. However, they do not seem to use abstractions for handling action failures, since they do not employ explicit representation for plans. In our approach agents often modify their plans at the lower levels in their plan structure, ....

David Chapman and Philip E. Agre. Abstract reasoning as emergent from concrete activity. In Workshop on Planning and Reasoning about Action, Portland, Oregon, June 1986. ERROR: undefined OFFENDING COMMAND: --------------C309E54E5EC23E7DF3C3092F-- STACK:

....to Deutsch s own work, particularly The Resolution of Conflict, 1973, chapters 7 and 8. CHAPTER 3. TRUST 30 agent making a model of another and the other s plans, and about the world. In situated action, the agent behaves in a completely reactive manner, thus models are not possible (Agre Chapman, 1987; Chapman Agre, 1987) and unnecessary. For Hobbes, everyone was in a state of war against all, thus the only model that was to be made concerned the potential harm to be had from the actions of others (Hobbes, 1946, chapter 13, see also Kavka, 1983) In more deliberative systems, a knowledge ....

....own work, particularly The Resolution of Conflict, 1973, chapters 7 and 8. CHAPTER 3. TRUST 30 agent making a model of another and the other s plans, and about the world. In situated action, the agent behaves in a completely reactive manner, thus models are not possible (Agre Chapman, 1987; Chapman Agre, 1987) and unnecessary. For Hobbes, everyone was in a state of war against all, thus the only model that was to be made concerned the potential harm to be had from the actions of others (Hobbes, 1946, chapter 13, see also Kavka, 1983) In more deliberative systems, a knowledge of others is much more ....

[Article contains additional citation context not shown here]

Chapman, David, & Agre, Philip E. 1987. Abstract Reasoning as Emergent from Concrete Activity. In: Georgeff, Mike P., & Lansky, Amy L. (eds), Reasoning about Actions and Plans. Morgan Kauffman.

....simple in itself this might appear, its power grows exponentially when several reactive components (behaviours) are connected and used within the same system. Interactions between such reactive components may result to what appears to an observer concrete activity or even purposeful reasoning (Chapman and Agre, 1986). The successes of the reactive approach have shown that it can elicit complex quasi intelligent appearing behaviour without resorting to use of any memory or model of the environment. Reactive systems sometimes exhibit competence on tasks that may previously have been deemed to require a world ....

Chapman, D. and P. Agre (1986). Abstract reasoning as emergent from concrete activity. In: M.P. Georgeff and A.L. Lansky (Eds.), Reasoning About Actions and Plans, Proceedings of the 1986 Workshop, Timberline, Oregon, Morgan Kaufmann, Los Altos, CA, 1986, pp. 411-424.

....to the sort of adversarial problem solving tasks studied by the Cognitivists. There are two argument for this claim. The first is somewhat controversial: the abstract reasoning needed to deal with an adversarial world is derived from concrete routine activity, rather than the other way around [CA86, Bee93] Whether this is true is a matter of some debate: Cognitivists claim it is false; the SA camp thinks it is true. The second argument for focusing on everyday life is pragmatic: if it is true that most of life is routine and unproblematic, then we can build systems that perform ....

D. Chapman and P. Agre. Abstract Reasoning as Emergent From Concrete Activity. In M. Georgeff and A. Lansky, editors, Reasoning about Actions and Plans. Morgan Kaufmann, 1986. Proceedings of the 1986 Timberline Workshop.

....his Master s thesis, in which he reported the theoretical difficulties with planning described above, and was coming to similar conclusions about the inadequacies of the symbolic AI model himself. Together with his co worker Agre, he began to explore alternatives to the AI planning paradigm (Chapman and Agre, 1986). Agre observed that most everyday activity is routine in the sense that it requires little if any new abstract reasoning. Most tasks, once learned, can be accomplished in a routine way, with little variation. Agre proposed that an efficient agent architecture could be based on the idea ....

Chapman, D. and Agre, P. (1986). Abstract reasoning as emergent from concrete activity. In Georgeff, M. P. and Lansky, A. L., editors, Reasoning About Actions & Plans --- Proceedings of the 1986 Workshop, pages 411--424. Morgan Kaufmann Publishers: San Mateo, CA.

....(skills, reflexes) a cognitive level (our sequencer) at between 10 seconds and 10 milliseconds, and a rational band (the planner) at greater than 10 seconds. But 3T also represents a smooth transition from event based reasoning to continuous action. This is somewhat different from earlier work (Chapman Agre 1986), which posits that abstract reasoning grows out of our embodied actions. Moreover, 3T s transition attempts to model the human trait of attending to critical local, concrete activity, but being able to reason more abstractly when the activity becomes routine. An example is driving a car. On a ....

Chapman, D., and Agre, P. E. 1986. Abstract reasoning as emergent from concrete activity. In Proceedings of the Workshop on Planning and Reasoning About Action.

....his Master s thesis, in which he reported the theoretical difficulties with planning described above, and was coming to similar conclusions about the inadequacies of the symbolic AI model himself. Together with his co worker Agre, he began to explore alternatives to the AI planning paradigm [25]. Agre observed that most everyday activity is routine in the sense that it requires little if any new abstract reasoning. Most tasks, once learned, can be accomplished in a routine way, with little variation. Agre proposed that an efficient agent architecture could be based on the idea ....

D. Chapman and P. Agre. Abstract reasoning as emergent from concrete activity. In M. P. Georgeff and A. L. Lansky, editors, Proceedings of the 1986 Workshop on Reasoning About Actions and Plans. Morgan Kaufmann Publishers, Inc., 1986.

....better, although there may be some limit to this improvement. In the next set of experiments we apply these design insights to a more difficult problem. The light as food problem Since we believe that abstract reasoning abilities, such as planning, arise developmentally from concrete activity (Chapman Agre, 1987), a connectionist, autonomous agent controller should also be able to exhibit plan like behavior if given a more complex problem. For this problem, a goal unit was added to the input layer. A positive value for the goal indicated that carbot should seek out the light (placed in one corner of the ....

Chapman, D. and Agre, P. E. 1987. Abstract reasoning as emergent from concrete activity. In Georgeff, M. P.

....identified above, Kaebling [62] describes an architecture for a reactive system for navigation in which a planner works incrementally with information obtained from sensor. Ambros Ingerson et al. 4] present a framework, IPEM, for integrating planning, execution and monitoring of the operations. [31] introduces execution and monitoring into TWEAK s partial plan representation . The control is carried out using a production system architecture with conflicts resolved by a scheduler. An example of Overseer Controller Sensors Observers vison force tactile Actuators Plant Disturbances Planner ....

D. Chapman and P. Agre. Abstract Reasoning as Emergent from Concrete Activity. In Reasoning about Actions and Plans. Morgan Kaufmann, 1987.

....to the sort of adversarial problem solving tasks studied by the cognitivists. There are two arguments for this claim. The first is somewhat controversial: the abstract reasoning needed to deal with an adversarial world is derived from concrete routine activity, rather than the other way around [6]. Whether this is true is a matter of some debate: Cognitivists claim it is false; the SA camp thinks it is true. The second argument for focusing on everyday life is pragmatic: if it is true that most of life is routine and unproblematic, then we can build systems that perform satisfactorily on a ....

Chapman, D.---Agre, P.: Abstract Reasoning as Emergent From Concrete Activity. In: M. Georgeff and A. Lansky, (Eds.) Reasoning about Actions and Plans. Morgan Kaufmann 1986. Proceedings of the 1986 Timberline Workshop.

....scheme) However simple in itself this might appear, its power grows exponentially when several reactive components are connected and used within the same system. Interactions between such reactive components may result to what appears to an observer concrete activity or even purposeful reasoning [5]. For the case of a robot, if S stands for its perceptions and R for the commands finally fed to the actuators, it appears as though something exists in between (S x R) This something is nothing else but a reactive algorithm. Reactive because as explained it does not maintain an explicit model ....

D. Chapman & P. Agre, Abstract reasoning as emergent from concrete activity, in M. P. Georgeff & A. L. Lansky (Eds.), Reasoning About Actions and Plans, Proceedings of the 1986 Workshop, Timberline, Oregon, Morgan Kaufmann, Los Altos, CA, 1986, pp. 411-424.

....his Master s thesis, in which he reported the theoretical difficulties with planning described above, and was coming to similar conclusions about the inadequacies of the symbolic AI model himself. Together with his co worker Agre, he began to explore alternatives to the AI planning paradigm [27]. Agre s work is conceptually more sophisticated than thats of Brooks. He observed that most everyday activity is routine in the sense that it requires little if any new abstract reasoning. Most tasks, once learned, can be accomplished in a routine way, with little variation. Agre ....

D. Chapman and P. Agre. Abstract reasoning as emergent from concrete activity. In M. P. Georgeff and A. L. Lansky, editors, Reasoning About Actions & Plans --- Proceedings of the 1986 Workshop, pages 411--424. Morgan Kaufmann Publishers: San Mateo, CA, 1986.

....1986b] and the system [Agre and Chapman, 1987] in particular. The system is specifically designed to be a robot programming language, and as such is defended by its ability to flexibly describe very general robotic tasks. The system is an implementation of a theory of situated activity , [Chapman and Agre, 1986] and is defended, not by its contribution to robot programming, but by the theory s power to describe realworld behavior. The system is both a programming language and a proposal for structuring behavior descriptions and can be defended on both counts. The representation syntax is clearly a robot ....

....4.1 and 4.9. 31 The form that a robot memory might take is heavily dependent on the type of information that can be expected from the robot s sensor systems. Two alternative sensor models explored in the planning literature are the fixed designator model, and the indexical functional model [Chapman and Agre, 1986, Agre and Chapman, 1987] The fixed designator model assumes that sensors can recognize individual items in the world and attach the same unique name to each whenever and wherever it is encountered. The indexical functional model assumes that individual items can be differentiated and classified, ....

[Article contains additional citation context not shown here]

David Chapman and Philip E. Agre. Abstract reasoning as emergent from concrete activity. In , Portland, Oregon, June 1986.

....workspaces that one occupies to perform a particular activity over a bounded period. In repairing a bicycle, for example, one might spread tools and bicycle parts about on the floor in patterns that have a cognitive significance in relationship to one s own body and cognitive and other states (Chapman Agre, 1986). One is not claiming this space as a permanent colony (it might be located on a patio in a Agre Horswill public park, for example) but one does lay claim to the space long enough to perform a customarily bounded task. ffl One s own private spaces: home, desk, office, car, trunks of stuff ....

Chapman, D., & Agre, P. E. (1986). Abstract reasoning as emergent from concrete activity.

No context found.

Chapman, D., and Agre, P. E., 1987. Abstract reasoning as emergent from concrete activity, Reasoning about Actions and Plans (M. P. Georgeff and A. L. Lansky, eds.), Los Altos, CA: Morgan Kaufmann, 411-424.

No context found.

D. Chapman and P. Agre. Abstract reasoning as emergent from concrete activity. In Workshop on Planning& Reasoning about Action, pages 251-266, Timberline OR, 1986.

No context found.

Chapman, D., and Agre, P., \Abstract Reasoning as Emergent from Concrete Activity" Proceedings, Workshop on Planning Timberline, OR, 1986, pp.251-266.

No context found.

D. Chapman and P. Agre. Abstract reasoning as emergent from concrete activity. In Workshop on Planning& Reasoning about Action, pages 251-266, Timberline OR, 1986.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC