The study of the acoustical in uence of the upstream (vocal-tract) airway in wind instrument performance has been a growing subject of interest in musical acoustics. In a recent study performed on trombone players, we observed a noticeable increase of the acoustic pressure near the fundamental frequency at the input of the vocal tract relative to the acoustic pressure at the input of the instru...

Identifying what problems can be solved in a given distributed system is a central question in distributed computing. In this series of works, we study this question in the context of asynchronous fault tolerant systems that can execute consensus. These systems can be those executing deterministic protocols with access to a consensus routine or those running randomized error-free protocols. A p...

In this paper, we present a new approach for programmable workflow composition. The workflow is represented as sequence of tasks with explicitly defined input/output data. Parallelism between tasks is implicitly defined by the data dependence. Users can create a more complicated workflow by scripts, including nested and parameterized workflows. The workflow will be executed in distributed

Many tasks in safety-critical embedded systems have hard real-time characteristics. AbsInt’s worst-case execution time analyzer aiT can estimate precise and safe upper bounds for the WCETs of program tasks, thus providing the basic input for verifying the real-time behavior of embedded applications.

Before 9 months, infants use sound to retrieve a stationary object hidden by darkness but not one hidden by occlusion, suggesting auditory input is more salient in the absence of visual input. This article addresses how audiovisual input affects 10-month-olds' search for displaced objects. In AB tasks, infants who previously retrieved an object at A subsequently fail to find it after it is disp...

Activation of the left midfusiform gyrus in response to reading words and pseudowords is such a reliable finding in functional imaging that this region has been called "the visual word form area" (VWFA). However, this label has recently been challenged, because activation in VWFA is also observed in other lexical tasks. We evaluated whether VWFA is necessary, sufficient, or specialized for read...

We consider the problem of training input-output recurrent neural networks (RNN) for sequence labeling tasks. We propose a novel spectral approach for learning the network parameters. It is based on decomposition of the cross-moment tensor between the output and a non-linear transformation of the input, based on score functions. We guarantee consistent learning with polynomial sample and comput...

Obtaining compact and discriminative features is one of the major challenges in many of the real-world image classification tasks such as face verification and object recognition. One possible approach is to represent input image on the basis of high-level features that carry semantic meaning which humans can understand. In this paper, a model coined deep attribute network (DAN) is proposed to ...

The objectives of this task are both to conduct research on seismic hazards, and to provide relevant input on the expected levels of these hazards to other tasks. Other tasks requiring this input include those dealing with inventory, fragility curves, rehabilitation strategies and demonstration projects. The corresponding input is provided in various formats depending on the intended use: eithe...

The objectives of this task are to conduct research on seismic hazards, and to provide relevant input on the expected levels of these hazards to other tasks. Other tasks requiring this input include those dealing with inventory, fragility curves, rehabilitation strategies and demonstration projects. The corresponding input is provided in various formats depending on the intended use: as peak gr...