Impairment of knowledge of unusual meanings of familiar words in brain damage and schizophrenia.

A test measuring knowledge of unusual meanings of familiar words was constructed, and was hypothesized to be more sensitive than traditional vocabulary tests to language impairment in schizophrenic and brain-damaged patients. In addition, the oft-noted inferiority of these patients on tests of abstract reasoning, as compared to vocabulary tests, was hypothesized to be a function of insensitivity of usual vocabulary tests. The new test provides evidence supporting both hypotheses in schizophrenic but not in brain-damaged patients. The results suggest that the similar outcome of impaired abstract reasoning in brain damage and schizophrenia may results from different processes in each disorder. (16 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Magnetomechanical detection of the specific activities of endonucleases by cantilevers

Thiolated nucleic acids 1 or 2 are immobilized on Au-coated cantilevers and hybridized with the complementary nucleic acids 1a or 2a associated with magnetic particles. The duplexes 1/1a or 2/2a include specific sequences for the scission by Apa I or Mse I, respectively. The cantilevers positioned in a flow cell are subjected to an external magnetic field, leading to the deflection of the cantilevers. Upon the specific scission of the DNA duplexes by Apa I or Mse I, the magnetic particles are disconnected from the cantilevers leading to their retraction to the rest position. The deflection/retraction of the cantilevers are followed by a conventional atomic force microscope optical detection system.     

Multi-wavelength all-optical networks with wavelengths outside theerbium-doped fiber amplifier bandwidth

Erbium-doped fiber amplifiers have become the dominating technology for signal amplification in all-optical networks. One constraint of EDFA's is that they have a much narrower bandwidth (≈25 nm) compared to the low-loss region (≈200 mn) of optical fiber. Instead of using only wavelengths within the bandwidth of EDFA's (i.e. the inband channels) for communication, we propose to include wavelengths outside the EDFA bandwidth (i.e. the outband channels) as well in order to increase the number of wavelengths and/or channel spacings that can be accommodated. Using outband wavelengths for sending messages presents a new constraint, namely that only if the power loss for transmitting a message is small enough can this message be transmitted on the outband wavelength. We develop wavelength-routing algorithms on arbitrary network topologies and wavelength assignments in hierarchical networks for sending messages subject to this constraint. We also analyze the SNR for inband/outband WDM signals     

Milepost GCC: Machine Learning Enabled Self-tuning Compiler

Tuning compiler optimizations for rapidly evolving hardware makes porting and extending an optimizing compiler for each new platform extremely challenging. Iterative optimization is a popular approach to adapting programs to a new architecture automatically using feedback-directed compilation. However, the large number of evaluations required for each program has prevented iterative compilation from widespread take-up in production compilers. Machine learning has been proposed to tune optimizations across programs systematically but is currently limited to a few transformations, long training phases and critically lacks publicly released, stable tools. Our approach is to develop a modular, extensible, self-tuning optimization infrastructure to automatically learn the best optimizations across multiple programs and architectures based on the correlation between program features, run-time behavior and optimizations. In this paper we describe Milepost GCC, the first publicly-available open-source machine learning-based compiler. It consists of an Interactive Compilation Interface (ICI) and plugins to extract program features and exchange optimization data with the cTuning.org open public repository. It automatically adapts the internal optimization heuristic at function-level granularity to improve execution time, code size and compilation time of a new program on a given architecture. Part of the MILEPOST technology together with low-level ICI-inspired plugin framework is now included in the mainline GCC. We developed machine learning plugins based on probabilistic and transductive approaches to predict good combinations of optimizations. Our preliminary experimental results show that it is possible to automatically reduce the execution time of individual MiBench programs, some by more than a factor of 2, while also improving compilation time and code size. On average we are able to reduce the execution time of the MiBench benchmark suite by 11% for the ARC reconfigurable processor. We also present a realistic multi-objective optimization scenario for Berkeley DB library using Milepost GCC and improve execution time by approximately 17%, while reducing compilation time and code size by 12% and 7% respectively on Intel Xeon processor.