V. Lesser, B. Horling, F. Klassner, A. Raja, T. Wagner, and S. Zhang. Big: An agent for resource-bounded information gathering and decision making. Artificial Intelligence, 118:197--244, 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....the degrees of such incompleteness or unsoundness must be a function of the available resources. Answers will often have to be approximate (where ideally, the reasoner can give us an indication of the quality of such approximations) Zilberstein and Russell, 1995, Russell et al. 1993, Lesser et al. 2000] Agent research leading KR Of course, not all of the above research challenges are new to KR, and many of them have been on the research agenda to some extent. Examples are approximate reasoning, trust, task independent formulation of knowledge, and reconciling multiple vocabularies. ....

Lesser, V., Hornling, B., Klassner, F., Raja, A., Wagner, T., and Zhang, S. (2000). Big: An agent for resource-bounded information gathering and decision making. Artificial Intelligence, 118(1-2):197--244. 3

....and replan, e#ciently reusing already obtained information, until a goal is satisfied or all ways of satisfying it fail. Planning has proven advantages in the task of information integration from multiple distributed sources; it hides from the user the process of data acquisition and manipulation [1, 10]. We take this idea further and weave such information integration into an ongoing human computer collaboration on a broader task that is the source of the information need. This setup creates advantages for both parties and thus results in more e#cient overall execution of the task. The user s ....

....paper has connections to work in several areas, most notably AI based collaborative interfaces, information integration systems and Internet search. Like many other information integration systems, Writer s Aid takes advantage of the breadth of bibliographic information available on the web. BIG [10] integrates several AI technologies, including resource bounded planning and scheduling to conduct an o#ine search for information on software packages based on a client s specification. Barish et al. 3] report on a query planning based system, called TheaterLoc, that searches online ....

V. Lesser, B. Horling, F. Klassner, A. Raja, T. Wagner, and S. Zhang. Big: An agent for resource-bounded information gathering and decision making. Artificial Intelligence, 118:197--244, 2000.

....article to ground the topics which are discussed and formulate examples. This section will brie y describe the environment and the particular challenges it o ers. Components of the SRTAarchitecture have also been used successfully in several other domains, suchasintelligent information gathering [11], intelligenthomecontrol [10] and supply chain [4] The distributed resource environment consists of several sensor nodes arranged in a region of nite area, as can be seen in Figure 1A. Each sensor node is autonomous, C B D Figure 1: High level distributed sensor allocation architecture. A) ....

....ways that goal may be achieved. The T MS structure can be generated in a varietyofways; in our case we use a T MS template library,whichwe use to dynamically instantiate and characterize structures to meet current conditions. Other options include generating the structure directly in code [11], or making use of an approximate base structure and then employing learning techniques to re ne it over time [8] The Design To Criteria component, used in the original controller described earlier, retains a critical role in SRTA. Where before it was responsible for both selecting an ....

Victor Lesser, Bryan Horling, Frank Klassner, Anita Raja, Thomas Wagner, and Shelley XQ. Zhang. BIG: An agent for resource-bounded information gathering and decision making. ######### ############, 118(12) :197-244, May2000.Elsevier Science Publishing.

....that considers the problem of quickly obtaining the first few answers to the query. WHIRL focuses on the important case where matching object names between different sources may require fuzzy matches, rather than exact matches. The BIG system, described in this issue s paper by Lesser et al. [76] addresses several additional issues related to information gathering, including the resource tradeoffs of different information gathering plans, extraction of data from unstructured sources and using the extracted data to further refine the search. A followup system to BIG is described by Grass ....

V. Lesser, B. Horling, F. Klassner, A. Raja, T. Wagner, S.XQ. Zhang, BIG: An agent for resource-bounded information gathering and decision making, Artificial Intelligence 118 (2000) 197--244 (this issue).

No context found.

Victor Lesser, Bryan Horling, Frank Klassner, Anita Raja, Thomas Wagner, and Shelley XQ. Zhang. BIG:an agent for resource-bounded information gathering and decision making. In Artificial Intelligence Journal, Special Issue on INternet Applications, 1999.

....to ground the topics which are discussed and formulate examples. This section will brie y describe the environment and the particular challenges it o ers. Components of the SRTA architecture have also been used successfully in several other domains, such as intelligent information gathering [22], intelligent home control [20] and supply chain [10] The distributed resource environment consists of several sensor nodes arranged in a region of nite area, as can be seen in Figure 1A. Each sensor node is autonomous, capable of communication, computation and observation through the attached ....

....Cache Check DTC Planner Cache Hit Method Usages Resource Used Fixed Schedule Results Schedule Merging Figure 2: High level agent control architecture. instantiate and characterize structures to meet current conditions. Other options include generating the structure directly in code [22], or making use of an approximate base structure and then employing learning techniques to re ne it over time [17] The Design To Criteria component, used in the original controller described earlier, retains a critical role in SRTA. Where before it was responsible for both selecting an ....

[Article contains additional citation context not shown here]

Victor Lesser, Bryan Horling, Frank Klassner, Anita Raja, Thomas Wagner, and Shelley XQ. Zhang. BIG: An agent for resource-bounded information gathering and decision making. Arti cial Intelligence, 118(12) :197-244, May 2000. Elsevier Science Publishing.

....execution path through a hierarchical task network such that the resultant schedule meets certain design criteria, such as real time deadlines, cost limits, and quality preferences. It is the heart of agent control in agent based systems such as the resource Bounded Information Gathering agent BIG [20] and the multi agent Intelligent Home [21] agent environment. Casting the language into an action selecting sequencing problem, the process is to select a subset of primitive actions from a set of candidate actions, and sequence them, so that the end result is an end to end schedule of an agent s ....

Victor Lesser, Bryan Horling, Frank Klassner, Anita Raja, Thomas Wagner, and Shelley XQ. Zhang. BIG:an agent for resource-bounded information gathering and decision making. In Artificial Intelligence Journal, Special Issue on INternet Applications, 1999.

....In the following sections we describe the components and discuss their strengths and weaknesses, and issues that have arisen during the course of our research. It is important to emphasize that this research pertains to complex problem solving agents, e. g, the BIG Information Gathering Agent [20] and [15, 4, 14] where the agents are situated in an environment, able to sense and effect, and have explicit representations of candidate tasks and explicit representations of different ways to go about performing the tasks. Additionally, tasks are quantified or have different performance ....

....fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and or a fee. Copyright 2000 ACM 0 89791 88 6 97 05 . 5.00 it via a model of the agent s problem solving process. The modeling framework we use is called TMS [6, 18] TMS task structures resemble complex and or graphs or HTNs, i.e. TMS is a hierarchical decomposition framework. Notable features of TMS models include the ....

[Article contains additional citation context not shown here]

Victor Lesser, Bryan Horling, Frank Klassner, Anita Raja, Thomas Wagner, and Shelley XQ. Zhang. BIG: An agent for resource-bounded information gathering and decision making. Artificial Intelligence, 118(1-2):197--244, May 2000. Elsevier Science Publishing.

....2000 ACM 0 89791 88 6 97 05 . 5.00 we describe our efforts to migrate our multi agent system framework into an environment which requires us to reason about and act in real time. Some or all of the technologies used in this framework have been used successfully in several other environments [10, 9]. They have not, however, been deployed in an environment demanding real time control that included real time coordination between agents. The particular environment we are operating under consists of several sensor nodes arranged in a region of finite area. Each sensor node is autonomous, ....

....we not only propagate uncertainty [21] but we can work to reduce it when important to the client. Until recently, DTC supplied online scheduling planning services to other components by being fast enough for the activities being scheduled. For example, in the BIG information gathering agent [10], scheduling planning accounted for less than 1 of the agent s execution time. However, in hard real time situations, being fast enough is not sufficient, as discussed in [19] The current generation scheduler supports hard real time deadlines at the grainsize afforded by the unix Linux operating ....

Victor Lesser, Bryan Horling, Frank Klassner, Anita Raja, Thomas Wagner, and Shelley XQ. Zhang. BIG: An agent for resource-bounded information gathering and decision making. Artificial Intelligence, 118(1-2):197--244, May 2000. Elsevier Science Publishing.

....Section 3, we discuss how scheduling a TMS network can be mapped into a finite horizon Markov Decision Process. However, to ground further discussion consider the TMS task structure shown in Figure 1. The task structure is a conceptual, simplified sub graph of a task structure emitted by the BIG [25] information gathering agent; it describes a portion of the information gathering process. The top level task is to construct product models of retail PC systems. It has two subtasks, Get Basic and Gather Reviews, both of which are decomposed into primitive actions, called methods, that are ....

....and Simulation) is a domain independent task modeling framework used to describe and reason about complex problem solving processes. TMS models are used in multi agent coordination research [11, 38] and are being used in many other research projects, including: Cooperative InformationGathering [27, 25], collaborative distributed design [12] distributed situation assessment [6] surviveable systems [35] multi agent diagnoses [3] intelligent environments [24] hospital patient scheduling [10] and coordination of software process [22] Typically a problem solver represents domain problem ....

[Article contains additional citation context not shown here]

V. Lesser, B. Horling, F. Klassner, A. Raja, T. Wagner, and S. XQ. Zhang. BIG: An agent for resource-bounded information gathering and decision making. To appear in the AIJ, 2000.

....and Simulation) 6] is a domain independent task modeling framework used to describe and reason about complex problem solving processes. TMS models are used in multi agent coordination research [24, 11] and are being used in many other research projects, including: cooperative informationgathering [14], hospital patient scheduling [5] intelligent environments [13] coordination of software process [12] and others [20] Typically, in our domain independent agent architecture, a domain specific problem solver or planner translates its problem solving options in TMS, possibly at some level of ....

....characteristics of primitive actions. These estimates can be refined over time through learning and reasoners typically replan and reschedule when unexpected events occur. To illustrate, consider Figure 2, which is a conceptual, simplified sub graph of a task structure emitted by the BIG [14] resource bounded information gathering agent; it describes a portion of the information gathering process. The top level task is to construct product models of retail PC systems. It has two subtasks, Get Basic and Gather Reviews, both of which are decomposed into actions, that are described in ....

Victor Lesser, Bryan Horling, Frank Klassner, Anita Raja, Thomas Wagner, and Shelley XQ. Zhang. BIG: An agent for resource-bounded information gathering and decision making. To appear in the AIJ, 2000.

....Simulation) 5] is a domain independent task modeling framework used to describe and reason about complex problem solving processes. TMS models are used in multi agent coordination research [24, 11] and are being used in many other research projects, including: cooperative information gathering [14, 15], collaborative distributed design [6] intelligent environments [13] coordination of software process [12] and others [21, 4] Typically, in our domain independent agent architecture, a domain specific problem solver or planner translates its problem solving options in TMS, possibly at some ....

....characteristics of primitive actions. These estimates can be refined over time through learning and reasoners typically replan and reschedule when unexpected events occur. To illustrate, consider Figure 2, which is a conceptual, simplified sub graph of a task structure emitted by the BIG [14, 15] resource bounded information gathering agent; it describes a portion of the information gathering process. The top level task is to construct product models of retail PC systems. It has two subtasks, Get Basic and Gather Reviews, both of which are decomposed into actions, that are described in ....

Victor Lesser, Bryan Horling, Frank Klassner, Anita Raja, Thomas Wagner, and Shelley XQ. Zhang. BIG: An agent for resource-bounded information gathering and decision making. To appear in the AIJ, 2000.

No context found.

V. Lesser, B. Horling, F. Klassner, A. Raja, T. Wagner, and S. Zhang. Big: An agent for resource-bounded information gathering and decision making. Artificial Intelligence, 118:197--244, 2000.

No context found.

V. Lesser, B. Horling, F. Klassner, A. Raja, T. Wagner, and S. Zhang. BIG: An agent for resource-bounded information gathering and decisionmaking. Artificial Intelligence Journal, Special Issue on Internet Information Agents, 118(1/2):197--244, 2000.

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