We discuss the modeling of count data whose empirical distribution is both multi-modal and overdispersed, and propose the Hermite distribution with covariates introduced through the conditional mean. The model is readily estimated by maximum likelihood, and nests the Poisson model as a special case. The Hermite regression model is applied to data for the number of banking and currency crises in...

DNN-based cross-modal retrieval is a research hotspot to retrieve across different modalities as image and text, but existing methods often face the challenge of insufficient cross-modal training data. In single-modal scenario, similar problem is usually relieved by transferring knowledge from largescale auxiliary datasets (as ImageNet). Knowledge from such single-modal datasets is also very us...

In this paper, we consider the influence of trust on the assimilation of acquired information into an agent’s belief. By use of modal logic tools, we characterize the relationship among belief, information acquisition and trust both semantically and axiomatically. The belief and information acquisition are respectively represented by KD45 and KD normal modal operators, whereas trust is expresse...

During a revisit to modal identification in the past three decades, an overview of the major developments and issues in modal identification is motivated. A brief review on the progress from Single-Input/Single-Output (SISO), Single-Input/Multi-Output (SIMO) to Multi-Input/Multi-Output (MIMO) identification in frequency, time and spatial domain, is presented. The latest developments and issues ...

It is known that the inconsistent distribution and representation of different modalities, such as image and text, cause the heterogeneity gap, which makes it very challenging to correlate such heterogeneous data. Recently, generative adversarial networks (GANs) have been proposed and shown its strong ability of modeling data distribution and learning discriminative representation, and most of ...

Presenting structural damage detection problem as an inverse model-updating approach is one of the well-known methods which can reach to informative features of damages. This paper proposes a model-based method for fault prognosis in engineering structures. A new damage-sensitive cost function is suggested by employing the main concepts of the Modal Assurance Criterion (MAC) on the first severa...

We study the data complexity of the modal query language MDatalog and its extension eMDatalog in basic modal logics. MDatalog is a modal extension of Datalog, while eMDatalog is the general modal Horn fragment with the allowedness condition. As the main results, we prove that the data complexity of MDatalog and eMDatalog in K4, KD4, and S4 is PSPACE-complete, in K is coNP-complete, and in KD , ...

The modal split model is one of the steps of the classical four-step travel demand planning. Predictive, descriptive, and prescriptive modal split models are essential to make a balance between travel demand and supply. To calibrate these models, it is necessary to detect and employ influential independent variables that are related to characteristics of travel modes, individual and family attr...

This paper investigates the dynamics of the engine driveline system for achieving accurate and online diagnostics of engine conditions using instantaneous angular speed (IAS). Based on a state space modelling approach and key structures of the system, a torsional vibration model is developed and its modal properties: modal frequency, damping ratio and shapes are evaluated. Then taking the outpu...