We introduce a method of incorporating additional knowledge sources into an HAM-based statistical acoustic model. The probabilistic relationship between information sources is first learned through a Bayesian network ...
详细信息
ISBN:
(纸本)9781605603162
We introduce a method of incorporating additional knowledge sources into an HAM-based statistical acoustic model. The probabilistic relationship between information sources is first learned through a Bayesian network to easily integrate any additional knowledge sources that might come from any domain and then the global joint probability density function (PDF) of the model is formulated. Where the model becomes too complex and direct BN inference is intractable, we utilize a junction tree algorithm to decompose the global joint PDF into a linked set of local conditional PDFs. This way, a simplified form of the model can be constructed and reliably estimated using a limited amount of training data. Here, we apply this framework to incorporate accents, gender, and wide-phonetic knowledge information at the HAM phonetic model level. The performance of the proposed method was evaluated on an LVCSR task using two different types of accented English speech data. Experimental results revealed that our method improves word accuracy with respect to standard HAM.
Releasing high-dimensional data enables a wide spectrum of data mining tasks. Yet, individual privacy has been a major obstacle to data sharing. In this paper, we consider the problem of releasing high-dimensional dat...
详细信息
ISBN:
(纸本)9781450336642
Releasing high-dimensional data enables a wide spectrum of data mining tasks. Yet, individual privacy has been a major obstacle to data sharing. In this paper, we consider the problem of releasing high-dimensional data with differential privacy guarantees. We propose a novel solution to preserve the joint distribution of a high-dimensional dataset. We first develop a robust sampling-based framework to systematically explore the dependencies among all attributes and subsequently build a dependency graph. This framework is coupled with a generic threshold mechanism to significantly improve accuracy. We then identify a set of marginal tables from the dependency graph to approximate the joint distribution based on the solid inference foundation of the junction tree algorithm while minimizing the resultant error. We prove that selecting the optimal marginals with the goal of minimizing error is NP-hard and, thus, design an approximation algorithm using an integer programming relaxation and the constrained concave-convex procedure. Extensive experiments on real datasets demonstrate that our solution substantially outperforms the state-of-the-art competitors.
Modem combat aircraft pilots increasingly rely on high-level fusion models (JDL Levels 2/3) to provide real-time engagement support in hostile situations. These models provide both Situational Awareness (SA) and Threa...
详细信息
ISBN:
(纸本)0819462918
Modem combat aircraft pilots increasingly rely on high-level fusion models (JDL Levels 2/3) to provide real-time engagement support in hostile situations. These models provide both Situational Awareness (SA) and Threat Assessment (TA) based on data and the relationships between the data. This information represents two distinct classes of uncertainty: vagueness and ambiguity. To address the needs associated with modeling both of these types of data uncertainty, an innovative hybrid approach was recently introduced, combining probability theory and possibility theory into a unified computational framework. The goal of this research is to qualitatively and quantitatively address the advantages and disadvantages of adopting this hybrid framework as well as identifying instances in which the combined model outperforms or is more appropriate than more classical inference approaches. To accomplish this task, domain specific models will be developed using different theoretical approaches and conventions, and then evaluated in comparison to situational ground truth to determine their accuracy and fidelity. Additionally, the performance tradeoff between accuracy and complexity will be examined in terms of computational cost to determine both the advantages and disadvantages of each approach.
One of the greatest challenges in modem combat is maintaining a high level of timely Situational Awareness (SA). In many situations, computational complexity and accuracy considerations make the development and deploy...
详细信息
ISBN:
(纸本)9780819466891
One of the greatest challenges in modem combat is maintaining a high level of timely Situational Awareness (SA). In many situations, computational complexity and accuracy considerations make the development and deployment of realtime, high-level inference tools very difficult. An innovative hybrid framework that combines Bayesian inference, in the form of Bayesian Networks, and Possibility Theory, in the form of Fuzzy Logic systems, has recently been introduced to provide a rigorous framework for high-level inference. In previous research, the theoretical basis and benefits of the hybrid approach have been developed. However, lacking is a concrete experimental comparison of the hybrid framework with traditional fusion methods, to demonstrate and quantify this benefit. The goal of this research, therefore, is to provide a statistical analysis on the comparison of the accuracy and performance of hybrid network theory, with pure Bayesian and Fuzzy systems and an inexact Bayesian system approximated using Particle Filtering. To accomplish this task, domain specific models will be developed under these different theoretical approaches and then evaluated, via Monte Carlo Simulation, in comparison to situational ground truth to measure accuracy and fidelity. Following this, a rigorous statistical analysis of the performance results will be performed, to quantify the benefit of hybrid inference to other fusion tools.
暂无评论