Results 11 - 20
of
81
Towards a formal approach to overhearing: Algorithms for conversation identification
- In Third International Conference on Autonomous Agents and Multiagent Systems (AAMAS-2004
, 2004
"... Overhearing is gaining attention as a generic method for cooperative monitoring of distributed, open, multiagent systems. It involves monitoring the routine conversations of agents–who know they are being overheard–to assist the agents, assess their progress, or suggest advice. While there have been ..."
Abstract
-
Cited by 17 (4 self)
- Add to MetaCart
(Show Context)
Overhearing is gaining attention as a generic method for cooperative monitoring of distributed, open, multiagent systems. It involves monitoring the routine conversations of agents–who know they are being overheard–to assist the agents, assess their progress, or suggest advice. While there have been several investigations of applications and methods of overhearing, no formal model of overhearing exists. This paper takes steps towards such a model. It first formalizes a conversation system–the set of conversations in a multi-agent system. It then defines a key step in overhearing–conversation recognition– identifying the conversations that took place within a system, given a set of overheard messages. We provide a skeleton algorithm for conversation recognition, and provide instantiations of it for settings involving no message loss, random message loss, and systematic message loss (such as always losing one side of the conversation). We analyze the complexity of these algorithms, and show that the systematic message loss algorithm, which is unique to overhearing, is significantly more efficient then the random loss algorithm (which is intractable). 1.
Towards monitoring of group interactions and social roles via overhearing
- In Proceedings of CIA-04
, 2004
"... Abstract. We are investigating how to provide intelligent, pervasive support of group of people within so-called “smart environments”. Our current main assumption, based on literature in psychology and organizational studies, is that a group performs some complex, routine task as a structured activi ..."
Abstract
-
Cited by 15 (1 self)
- Add to MetaCart
(Show Context)
Abstract. We are investigating how to provide intelligent, pervasive support of group of people within so-called “smart environments”. Our current main assumption, based on literature in psychology and organizational studies, is that a group performs some complex, routine task as a structured activity, that is, by following some protocols that allow its members to coordinate and share a common understanding about the current progress towards the group’s goal and the roles currently played by each member. If this is the case, a condition to provide support to a group activity by artificial agents is to share the same understanding. To this end, we have identified two initial goals: first, being able to understand if a group activity is progressing with respect to its expected evolution, by analyzing what is happening within the smart environment; second, recognizing what are the social roles of the group members, taking in mind that these are not necessarily pre-assigned and may change in time. This paper sketches a preliminary approach to these issues and a computational model for an overhearer agent. We suggest a preliminary set of rules for conversation analysis and social role recognition, and validate them against the simple case of implicit organizations, which – being artificial – follow well-known protocols.
Model-Free Execution Monitoring in Behavior-Based Mobile Robotics
- in: Proceedings of the International Conference on Advanced Robotics (ICAR
, 2003
"... In this paper we present a model-free execution monitor for behavior-based mobile robots. By modelfree we mean that the monitoring is based only on the actual execution, without involving any predictive models of the controlled system. Model-free monitors are especially suitable for systems where it ..."
Abstract
-
Cited by 14 (2 self)
- Add to MetaCart
(Show Context)
In this paper we present a model-free execution monitor for behavior-based mobile robots. By modelfree we mean that the monitoring is based only on the actual execution, without involving any predictive models of the controlled system. Model-free monitors are especially suitable for systems where it is hard to obtain adequate models. In our approach we analyze the activation levels of the di#erent behaviors using a pattern recognition technique. Our model-free execution monitor, which is realized by radial basis function networks, is shown to give a high performance in several realistic simulations.
Conflicts in teamwork: Hybrids to the rescue
- In AAMAS ’05: Proceedings of the fourth international
, 2005
"... Today within the AAMAS community, we see at least four competing approaches to building multiagent systems: beliefdesire-intention (BDI), distributed constraint optimization (DCOP), distributed POMDPs, and auctions or game-theoretic approaches. While there is exciting progress within each approach, ..."
Abstract
-
Cited by 12 (4 self)
- Add to MetaCart
(Show Context)
Today within the AAMAS community, we see at least four competing approaches to building multiagent systems: beliefdesire-intention (BDI), distributed constraint optimization (DCOP), distributed POMDPs, and auctions or game-theoretic approaches. While there is exciting progress within each approach, there is a lack of cross-cutting research. This paper highlights hybrid approaches for multiagent teamwork. In particular, for the past decade, the TEAMCORE research group has focused on building agent teams in complex, dynamic domains. While our early work was inspired by BDI, we will present an overview of recent research that uses DCOPs and distributed POMDPs in building agent teams. While DCOP and distributed POMDP algorithms provide promising results, hybrid approaches help us address problems of scalability and expressiveness. For example, in the BDI-POMDP hybrid approach, BDI team plans are exploited to improve POMDP tractability, and POMDPs improve BDI team plan performance. We present some recent results from applying this approach in a Disaster Rescue simulation domain being developed with help from the Los
Location-based reasoning about complex multiagent behavior
- In Journal of Artificial Intelligence Research. AI Access Foundation
, 2011
"... Recent research has shown that surprisingly rich models of human activity can be learned from GPS (positional) data. However, most effort to date has concentrated on modeling single individuals or statistical properties of groups of people. Moreover, prior work focused solely on modeling actual succ ..."
Abstract
-
Cited by 11 (3 self)
- Add to MetaCart
(Show Context)
Recent research has shown that surprisingly rich models of human activity can be learned from GPS (positional) data. However, most effort to date has concentrated on modeling single individuals or statistical properties of groups of people. Moreover, prior work focused solely on modeling actual successful executions (and not failed or attempted executions) of the activities of interest. We, in contrast, take on the task of understanding human interactions, attempted interactions, and intentions from noisy sensor data in a fully relational multi-agent setting. We use a real-world game of capture the flag to illustrate our approach in a well-defined domain that involves many distinct cooperative and competitive joint activities. We model the domain using Markov logic, a statistical-relational language, and learn a theory that jointly denoises the data and infers occurrences of high-level activities, such as a player capturing an enemy. Our unified model combines constraints imposed by the geometry of the game area, the motion model of the players, and by the rules and dynamics of the game in a probabilistically and logically sound fashion. We show that while it may be impossible to directly detect a multi-agent activity due to sensor noise or malfunction, the occurrence of the activity can still be inferred by considering both its impact on the
Monitoring Agents using Declarative Planning
- In LNAI 2821
, 2003
"... Abstract. We consider the following problem: Given a particular description of a multi-agent system (MAS), is it implemented properly? We assume we are given (possibly incomplete) information and aim at refuting that the given system is implemented properly. In our approach, agent collaboration is d ..."
Abstract
-
Cited by 10 (2 self)
- Add to MetaCart
(Show Context)
Abstract. We consider the following problem: Given a particular description of a multi-agent system (MAS), is it implemented properly? We assume we are given (possibly incomplete) information and aim at refuting that the given system is implemented properly. In our approach, agent collaboration is described as an action theory. Action sequences reaching the collaboration goal are computed by a planner, whose compliance with the actual MAS behaviour allows to detect possible collaboration failures. The approach can be fruitfully applied to aid offline testing of a MAS implementation,
Towards Dynamic Tracking of Multi-Agents Teams: An Initial Report
, 2007
"... This paper takes first steps to address the challenge of plan recognition for dynamic multi-agents teams, in the context of suspicious behavior recognition. Plan recognition is the process of inferring other agents ’ plans and goals based on their observable actions. Team plan recognition poses the ..."
Abstract
-
Cited by 10 (1 self)
- Add to MetaCart
This paper takes first steps to address the challenge of plan recognition for dynamic multi-agents teams, in the context of suspicious behavior recognition. Plan recognition is the process of inferring other agents ’ plans and goals based on their observable actions. Team plan recognition poses the challenge of such inference, of a team’s joint goals and plans. Most previous work have focused on recognizing specific (and limited) coordinated behaviors and do not deal with the problem of identifying interactions between groups of agents, and with identifying suspicious behavior from this information. In contrast, this paper utilizes the information from group of agents, to identify the interactions between groups of agents, using a Dynamic Hierarchical Group Model (DHGM) that tracks the dynamic grouping and ungrouping of agents. We show how such information can be used to identify potential suspicious behavior. These suspicious behaviors can be captured only when tracking individuals with respect to the group and not as individuals. For example, identifying passenger in the airport that behaves differently from other passengers in the same group. While reasoning about individual agents in a multi-agents framework is expensive, we reduce this complexity by utilizing the DHGM that encapsulate shared data of agents in the same group.
Monitoring and Organizational-Level Adaptation of Multi-Agent Systems
- In Proc. of AAMAS’04
, 2004
"... Static organizational structures are not always suited for large-scale open multi-agent systems. On the other hand, emergent organizational structures may lead to undesirable behavior despite agent-level adaptation. We thus propose a new adaptive multi-agent architecture with both agent-level and or ..."
Abstract
-
Cited by 10 (1 self)
- Add to MetaCart
Static organizational structures are not always suited for large-scale open multi-agent systems. On the other hand, emergent organizational structures may lead to undesirable behavior despite agent-level adaptation. We thus propose a new adaptive multi-agent architecture with both agent-level and organization-level adaptation. The global-level adaptation is based on the monitoring of the system's behavior and the dynamic reification of an organizational structure. This structure is used to prevent or detect undesirable behavior and take the required corrective actions. This architecture is applied to fault-tolerant multi-agent systems.
Oversensing with a softbody in the environment: Another dimension of observation
- Proceedings of Modeling Others from Observation at International Joint Conference on Artificial Intelligence
, 2005
"... Research in Multi-Agent Systems extensively approached observation mechanisms from outside. Agents are black boxes and much work allows inferring internals or collective activities from the observation of their interactions. In this paper, we propose to extend the expressiveness of software agents a ..."
Abstract
-
Cited by 9 (1 self)
- Add to MetaCart
Research in Multi-Agent Systems extensively approached observation mechanisms from outside. Agents are black boxes and much work allows inferring internals or collective activities from the observation of their interactions. In this paper, we propose to extend the expressiveness of software agents and enact new observable features. Our framework leads to an extended definition of agent; the central role of a software environment to deal with these agents; and a mechanism we named oversensing to refer to the observation of these agents. This paper presents these notions and provides preliminary insights on their exploitation. 1
Towards a monitoring framework for agent-based contract systems
- In: CIA ’08: Proceedings of the 12th international workshop on Cooperative Information Agents XII
"... Abstract. The behaviours of autonomous agents may deviate from those deemed to be for the good of the societal systems of which they are a part. Norms have therefore been proposed as a means to regulate agent behaviours in open and dynamic systems, and may be encoded in electronic contracts in order ..."
Abstract
-
Cited by 9 (4 self)
- Add to MetaCart
(Show Context)
Abstract. The behaviours of autonomous agents may deviate from those deemed to be for the good of the societal systems of which they are a part. Norms have therefore been proposed as a means to regulate agent behaviours in open and dynamic systems, and may be encoded in electronic contracts in order to specify the obliged, permitted and prohibited behaviours of agents that are signatories to such contracts. Enactment and management of electronic contracts thus enables the use of regulatory mechanisms to ensure that agent behaviours comply with the encoded norms. To facilitate such mechanisms requires monitoring in order to detect and explain violation of norms. In this paper we propose a framework for monitoring that is to be implemented and integrated into a suite of contract enactment and management tools. The framework adopts a non-intrusive approach to monitoring, whereby the states of a contract with respect to its contained norms can be inferred on the basis of messages exchanged. Specifically, the framework deploys agents that observe messages sent between contract signatories, where these messages correspond to agent behaviours and therefore indicate whether norms are, or are in danger of, being violated. 1
