Carnegie Mellon University

About: Carnegie Mellon University is a education organization based out in Pittsburgh, Pennsylvania, United States. It is known for research contribution in the topics: Computer science & Robot. The organization has 36317 authors who have published 104359 publications receiving 5975734 citations. The organization is also known as: CMU & Carnegie Mellon.

The child language data exchange system.

Abstract: The study of language acquisition underwent a major revolution in the late 1950s as a result of the dissemination of technology permitting high-quality tape-recording of children in the family setting. This new technology led to major breakthroughs in the quality of both data and theory. The field is now at the threshold of a possible second major breakthrough stimulated by the dissemination of personal computing. Researchers are now able to transcribe tape-recorded data into computer files. With this new medium it is easy to conduct global searches for word combinations across collections of files. It is also possible to enter new codings of the basic text line. Because of the speed and accuracy with which computer files can be copied, it is now much easier to share data between researchers. To foster this sharing of computerized data, a group of child language researchers has established the Child Language Data Exchange System (CHILDES). This article details the formation of the CHILDES, the governance of the system, the nature of the database, the shape of the coding conventions, and the types of computer programs being developed.

Transport of molecules across tumor vasculature.

Abstract: The vascular-extravascular exchange of fluid and solute molecules in a tissue is determined by three transport parameters (vascular permeability, P, hydraulic conductivity, Lp, and reflection coefficient, σ); the surface area for exchange, A; and the transluminal concentration and pressure gradients. The transport parameters and the exchange area for a given molecule are governed by the structure of the vessel wall. In general, tumor vessels have wide interendothelial junctions; large number of fenestrae and transendothelial channels formed by vesicles; and discontinuous or absent basement membrane. While these factors favor movement of molecules across tumor vessels, high interstitial pressure and low microvascular pressure may retard extravasation of molecules and cells, especially in large tumors. These characteristics of the transvascular transport have significant implications in tumor growth, metastasis, detection and treatment.

Clearing the clouds: a study of emerging scale-out workloads on modern hardware

Abstract: Emerging scale-out workloads require extensive amounts of computational resources. However, data centers using modern server hardware face physical constraints in space and power, limiting further expansion and calling for improvements in the computational density per server and in the per-operation energy. Continuing to improve the computational resources of the cloud while staying within physical constraints mandates optimizing server efficiency to ensure that server hardware closely matches the needs of scale-out workloads.In this work, we introduce CloudSuite, a benchmark suite of emerging scale-out workloads. We use performance counters on modern servers to study scale-out workloads, finding that today's predominant processor micro-architecture is inefficient for running these workloads. We find that inefficiency comes from the mismatch between the workload needs and modern processors, particularly in the organization of instruction and data memory systems and the processor core micro-architecture. Moreover, while today's predominant micro-architecture is inefficient when executing scale-out workloads, we find that continuing the current trends will further exacerbate the inefficiency in the future. In this work, we identify the key micro-architectural needs of scale-out workloads, calling for a change in the trajectory of server processors that would lead to improved computational density and power efficiency in data centers.

Unsupervised Visual Representation Learning by Context Prediction

Abstract: This work explores the use of spatial context as a source of free and plentiful supervisory signal for training a rich visual representation. Given only a large, unlabeled image collection, we extract random pairs of patches from each image and train a convolutional neural net to predict the position of the second patch relative to the first. We argue that doing well on this task requires the model to learn to recognize objects and their parts. We demonstrate that the feature representation learned using this within-image context indeed captures visual similarity across images. For example, this representation allows us to perform unsupervised visual discovery of objects like cats, people, and even birds from the Pascal VOC 2011 detection dataset. Furthermore, we show that the learned ConvNet can be used in the R-CNN framework and provides a significant boost over a randomly-initialized ConvNet, resulting in state-of-the-art performance among algorithms which use only Pascal-provided training set annotations.