Adaptive Dialogue Strategy Selection through Imprecise Probabilistic Query Answering

In a human-computer dialogue system, the dialogue strategy can range from very restrictive to highly flexible. Each specific dialogue style has its pros and cons and a dialogue system needs to select the most appropriate style for a given user. During the course of interaction, the dialogue style can change based on a user’s response and the system observation of the user. This allows a dialogue system to understand a user better and provide a more suitable way of communication. Since measures of the quality of the user’s interaction with the system can be incomplete and uncertain, frameworks for reasoning with uncertain and incomplete information can help the system make better decisions when it chooses a dialogue strategy. In this paper, we investigate how to select a dialogue strategy based on aggregating the factors detected during the interaction with the user. For this purpose, we use probabilistic logic programming (PLP) to model probabilistic knowledge about how these factors will affect the degree of freedom of a dialogue. When a dialogue system needs to know which strategy is more suitable, an appropriate query can be executed against the PLP and a probabilistic solution with a degree of satisfaction is returned. The degree of satisfaction reveals how much the system can trust the probability attached to the solution.

Preliminary analysis of organic Rankine cycles to improve vehicle efficiency

This paper presents the background rationale and key findings for a model-based study of supercritical waste heat recovery organic Rankine cycles. The paper’s objective is to cover the necessary groundwork to facilitate the future operation of a thermodynamic organic Rankine cycle model under realistic thermodynamic boundary conditions for performance optimisation of organic Rankine cycles. This involves determining the type of power cycle for organic Rankine cycles, the circuit configuration and suitable boundary conditions. The study focuses on multiple heat sources from vehicles but the findings are generally applicable, with careful consideration, to any waste heat recovery system. This paper introduces waste heat recovery and discusses the general merits of organic fluids versus water and supercritical operation versus subcritical operation from a theoretical perspective and, where possible, from a practical perspective. The benefits of regeneration are investigated from an efficiency perspective for selected subcritical and supercritical conditions. A simulation model is described with an introduction to some general Rankine cycle boundary conditions. The paper describes the analysis of real hybrid vehicle data from several driving cycles and its manipulation to represent the thermal inertia for model heat input boundary conditions. Basic theory suggests that selecting the operating pressures and temperatures to maximise the Rankine cycle performance is relatively straightforward. However, it was found that this may not be the case for an organic Rankine cycle operating in a vehicle. When operating in a driving cycle, the available heat and its quality can vary with the power output and between heat sources. For example, the available coolant heat does not vary much with the load, whereas the quantity and quality of the exhaust heat varies considerably. The key objective for operation in the vehicle is optimum utilisation of the available heat by delivering the maximum work out. The fluid selection process and the presentation and analysis of the final results of the simulation work on organic Rankine cycles are the subjects of two future publications.

Game-­theoretic Resource Allocation with Real-­time Probabilistic Surveillance Information

Game-theoretic security resource allocation problems have generated significant interest in the area of designing and developing security systems. These approaches traditionally utilize the Stackelberg game model for security resource scheduling in order to improve the protection of critical assets. The basic assumption in Stackelberg games is that a defender will act first, then an attacker will choose their best response after observing the defender’s strategy commitment (e.g., protecting a specific asset). Thus, it requires an attacker’s full or partial observation of a defender’s strategy. This assumption is unrealistic in real-time threat recognition and prevention. In this paper, we propose a new solution concept (i.e., a method to predict how a game will be played) for deriving the defender’s optimal strategy based on the principle of acceptable costs of minimax regret. Moreover, we demonstrate the advantages of this solution concept by analyzing its properties.

Uncertainty in the Cloud: An Angel-Daemon Approach to Modelling Performance

Uncertainty profiles are used to study the effects of contention within cloud and service-based environments. An uncertainty profile provides a qualitative description of an environment whose quality of service (QoS) may fluctuate unpredictably. Uncertain environments are modelled by strategic games with two agents; a daemon is used to represent overload and high resource contention; an angel is used to represent an idealised resource allocation situation with no underlying contention. Assessments of uncertainty profiles are useful in two ways: firstly, they provide a broad understanding of how environmental stress can effect an application’s performance (and reliability); secondly, they allow the effects of introducing redundancy into a computation to be assessed

The Robustness of Periodic Orchestrations in Uncertain Evolving Environments

A framework for assessing the robustness of long-duration repetitive orchestrations in uncertain evolving environments is proposed. The model assumes that service-based evaluation environments are stable over short time-frames only; over longer periods service-based environments evolve as demand fluctuates and contention for shared resources varies. The behaviour of a short-duration orchestration E in a stable environment is assessed by an uncertainty profile U and a corresponding zero-sum angel-daemon game Γ(U) [2]. Here the angel-daemon approach is extended to assess evolving environments by means of a subfamily of stochastic games. These games are called strategy oblivious because their transition probabilities are strategy independent. It is shown that the value of a strategy oblivious stochastic game is well defined and that it can be computed by solving a linear system. Finally, the proposed stochastic framework is used to assess the evolution of the Gabrmn IT system.

Learning Bounded Tree-Width Bayesian Networks via Sampling

Learning Bayesian networks with bounded tree-width has attracted much attention recently, because low tree-width allows exact inference to be performed efficiently. Some existing methods [12, 14] tackle the problem by using k-trees to learn the optimal Bayesian network with tree-width up to k. In this paper, we propose a sampling method to efficiently find representative k-trees by introducing an Informative score function to characterize the quality of a k-tree. The proposed algorithm can efficiently learn a Bayesian network with tree-width at most k. Experiment results indicate that our approach is comparable with exact methods, but is much more computationally efficient.

Approximating Credal Network Inferences by Linear Programming

An algorithm for approximate credal network updating is presented. The problem in its general formulation is a multilinear optimization task, which can be linearized by an appropriate rule for fixing all the local models apart from those of a single variable. This simple idea can be iterated and quickly leads to very accurate inferences. The approach can also be specialized to classification with credal networks based on the maximality criterion. A complexity analysis for both the problem and the algorithm is reported together with numerical experiments, which confirm the good performance of the method. While the inner approximation produced by the algorithm gives rise to a classifier which might return a subset of the optimal class set, preliminary empirical results suggest that the accuracy of the optimal class set is seldom affected by the approximate probabilities