John S Breese - Academia.edu (original) (raw)

Papers by John S Breese

Embodied Conversational Agents, 2000

... settings, some of the most reliable measures of emotional state involve physiological sensing... more ... settings, some of the most reliable measures of emotional state involve physiological sensing such ... example," you should definitely" versus" perhaps you might like to." Our approach to differentiating ... the likelihood that an individual will choose to speak in a positive (judgmental ...

Machine Intelligence and Pattern Recognition, 1990

This paper presents an approach to the design of autonomous, real-time systems operating in uncer... more This paper presents an approach to the design of autonomous, real-time systems operating in uncertain environments. We address issues of problem solving and reflective control of reasoning under uncertainty in terms of two fundamental elements: 1) a set of decision-theoretic models for selecting among alternative problem-solving methods and 2) a general computational architecture for resource-bounded problem solving. The decision-theoretic models provide a set of principles for choosing among alternative problem-solving methods based on their relative costs and benefits, where benefits are characterized in terms of the value of information provided by the output of a reasoning activity. The output may be an estimate of some uncertain quantity or a recommendation for action. The computational architecture, called Schemer-II, provides for interleaving of and communication among various problem-solving subsystems. These subsystems provide alternative approaches to information gathering, belief refinement, solution construction, and solution execution. In particular, the architecture provides a mechanism for interrupting the subsystems in response to critical events. We provide a decision theoretic account for scheduling problem-solving elements and for critical-event-driven interruption of activities in an architecture such as Schemer-II.

Eprint Arxiv 1302 4932, Feb 20, 2013

We describe an application of belief networks to the diagnosis of bottlenecks in computer systems... more We describe an application of belief networks to the diagnosis of bottlenecks in computer systems. The technique relies on a high-level functional model of the interaction between application workloads, the Windows NT operating system, and system hardware. Given a workload description, the model predicts the values of observable system counters available from the Windows NT performance monitoring tool. Uncertainty in workloads, predictions, and counter values are characterized with Gaussian distributions. During diagnostic inference, we use observed performance monitor values to find the most probable assignment to the workload parameters. In this paper we provide some background on automated bottleneck detection, describe the structure of the system model, and discuss empirical procedures for model calibration and verification. Part of the calibration process includes generating a dataset to estimate a multivariate Gaussian error model. Initial results in diagnosing bottlenecks are presented.

Computational Intelligence, 1992

We describe the structure of a Bayesian network designed to monitor the behavior of a user intera... more We describe the structure of a Bayesian network designed to monitor the behavior of a user interacting with a conversational computer and use that information to estimate the user's emotional state. Our model of emotional state uses two discrete dimensions, valence (bad to good) and arousal (calm to excited), to capture the dominant aspects of physical emotional response. Emotion is

Distributed Artificial Intelligence, 1989

Abstract Recent work on interactions among rational agents has put forward a computationally trac... more Abstract Recent work on interactions among rational agents has put forward a computationally tractable, deduction-based scheme for automated agents to use in analyzing multiagent encounters. While the theory has defined irrational actions, it has underconstrained an agent's choices: there are many situations where an agent in the previous framework was faced with several potentially rational actions, and no way of choosing among them. This paper presents a probabilistic extension to the previous framework of Genesereth, Ginsberg, and Rosenschein [Genesereth et al. 1986] that provides agents with a mechanism for further refining their choice of rational moves. At the same time, it maintains the computational attractiveness of the previous approach. The probabilistic extension is explicitly representing uncertainty about other players' moves. A three-level hierarchy of rationality is defined, corresponding to ordinal, stochastic, and utility dominance among alternative outcomes. The previous deduction-based formalism is recast in probabilistic terms and is seen to be a particular special case of a more encompassing dominance theory. A technique is presented for using the dominance ideas in interactions with other agents operating under various types of rationality.

Machine Intelligence and Pattern Recognition, 1990

We describe a mechanism for performing probabilistic reasoning in influence diagrams us ing inter... more We describe a mechanism for performing probabilistic reasoning in influence diagrams us ing interval rather than point valued probabilities. We derive the procedures for node removal (corresponding to conditional expectation) and arc reversal (corresponding to Bayesian condi tioning) in influence diagrams where lower bounds on probabilities are stored at each node. The resulting bounds for the transformed diagram are shown to be optimal within the class of constraints on probability distributions which can be expressed exclusively as lower bounds on the component probabilities of the diagram. Sequences of these operations can be performed to answer probabilistic queries with indeterminacies in the input and for performing sensitivity analysis on an influence diagram. The storage requirements and computational complexity of this approach are comparable to those for point-valued probabilistic inference mechanisms, mak ing the approach attactive for performing sensitivity analysis and where probability information is not available. Limited empirical data on an implementation of the methodology is provided.

We describe a framework for constructing,a model of emotions and personality for a computational ... more We describe a framework for constructing,a model of emotions and personality for a computational agent. The architecture uses dynamic models of emotions and personality encoded as Bayesian networks to 1) diag- nose the emotions and personality of the user, and 2) generate appropriate behavior by an automated agent. Classes of interaction that are interpreted and/or gen- erated include such things as choice of wording, char- acteristics of speech (speed and pitch), gesture, and facial expression. In particular, we describe the struc-

The Knowledge Engineering Review, 1992

In recent years there has been a growing interest among AI researchers in probabilistic and decis... more In recent years there has been a growing interest among AI researchers in probabilistic and decision modelling, spurred by significant advances in representation and computation with network modelling formalisms. In applying these techniques to decision support tasks, fixed network models have proven to be inadequately expressive when a broad range of situations must be handled. Hence many researchers have sought to combine the strengths of flexible knowledge representation languages with the normative status and well-understood computational properties of decision-modelling formalisms and algorithms. One approach is to encode general knowledge in an expressive language, then dynamically construct a decision model for each particular situation or problem instance. We have developed several systems adopting this approach, which illustrate a variety of interesting techniques and design issues.

IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans, 1996

A Bayesian network is a probabilistic representation for uncertain relationships, which has prove... more A Bayesian network is a probabilistic representation for uncertain relationships, which has proven to be useful for modeling real-world problems. When there are many potential causes of a given effect, however, both probability assessment and inference using a Bayesian network can be difficult. In this paper, we describe causal independence, a collection of conditional independence assertions and functional relationships that are often appropriate to apply to the representation of the uncertain interactions between causes and effect. We show how the use of causal independence in a Bayesian network can greatly simplify probability assessment as well as probabilistic inference.

Proceedings of the Sixth Conference on …, 1990

212 in Proceedings of the Sixth Conference on Uncertainty in Artificial Intelligence, Cambridge, ... more 212 in Proceedings of the Sixth Conference on Uncertainty in Artificial Intelligence, Cambridge, MA, August, 1990. IDEAL: A software package for analysis of influence diagrams Sampath Srinivas srinivas@rpal.com Jack Bréese breese@rpal.com Rockwell Science ...

By reading, you can know the knowledge and things more, not only about what you get from people t... more By reading, you can know the knowledge and things more, not only about what you get from people to people. Book will be more trusted. As this uai 01 proceedings of the 17th conference in uncertainty in artificial intelligence, it will really give you the good idea to be successful. It is not only for you to be success in certain life you can be successful in everything. The success can be started by knowing the basic knowledge and do actions.

We introduce and analyze the pro blem of the compilation of decision models from a. decision-theo... more We introduce and analyze the pro blem of the compilation of decision models from a. decision-theoretic perspective. The techniques described allow us to evaluate various configurations of compiled knowledge given the nature of evidential relationships in a. domain, the utilities associated with alternative actions, the costs of run-time delays, and the costs of memory. We describe procedures for selecting a subset of the total observations available to be incorporated into a. compiled situation-action mapping, in the context of a binary decision with conditional independence of evidence. The methods allow us to incrementally select the best pieces of evidence to add to the set of compiled knowledge in an engineering setting. After presenting several approaches to compilation, we exercise one of the methods to provide insight into the relationship between the distribution over weights of evidence and the preferred degree of compilation.

We develop a series of approximations for decision-theoretic troubleshooting under uncertainty. O... more We develop a series of approximations for decision-theoretic troubleshooting under uncertainty. Our approach generates troubleshooting plans in the face of uncertainty in the relationships among components and device status, observations, as well as the affect of actions on device status. Included in our approach is a Bayesian-network representation of these relationships. We have applied our technique successfully to troubleshooting problems with printing, photocopier feeders, automobiles, and gas turbines. We report empirical findings demonstrating the high quality of plans produced by our approach.

We have developed an expert system for diagnosis of efficiency problems for large gas turbines. T... more We have developed an expert system for diagnosis of efficiency problems for large gas turbines. The system relies on a model-based approach that combines an expert's probabilistic assessments with statistical data and thermodynamic analysis. The system employs a causal probabilistic graph, called a belief network, to update the likelihoods of alternative faults given information about diverse classes of information. In response to any subset of findings or reported observations, the system suggests the most cost-effective tests to perform to determine the source of a performance problem. We discuss the decision-analytic methodology that underlies the development of the system and present results of an initial version of the system. Finally, we discuss future planned development and evaluation, toward the ultimate goal of applying the system in the day-today maintenance of gasturbine power plants.

Embodied Conversational Agents, 2000

Machine Intelligence and Pattern Recognition, 1990

Eprint Arxiv 1302 4932, Feb 20, 2013

Computational Intelligence, 1992

Distributed Artificial Intelligence, 1989

Machine Intelligence and Pattern Recognition, 1990

The Knowledge Engineering Review, 1992

IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans, 1996

Proceedings of the Sixth Conference on …, 1990