A simple network model simulates hippocampal place fields: II. Computing goal-directed trajectories and memory fields.

Place cells have been described as the computational elements of a neuronal cognitive mapping system that encodes and stores relationships among spatial stimuli (J. O'Keefe and L. Nadel, 1978). Furthermore, place cells seem to encode remembered locations because neural activity is maintained when the visual stimuli that influence place field location are vastly degraded, such as when cues are removed or the lights are turned off (O'Keefe and A. Speakman, 1987; G. J. Quirk et al, 1990). A feed-forward network model that mapped visual input onto a representation of location simulated some basic properties of hippocampal place fields, including resistance to disruption after partial cue memory (M. L. Shapiro and P. A. Hetherington, 1993). However, the simulated place fields required visual input for their activation. It is reported that a network that incorporates feedback (1) computed correct trajectories toward simulated goals and (2) simulated place fields that persist in the absence of visual input. The simulation suggests that feedback properties can provide a computational account of O'Keefe and Speakman's data. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Recovered memories in theory and practice.

The final report of the American Psychological Association Working Group on Investigation of Memories of Childhood Abuse (J. L. Alpert et al, see record 2000-13581-002) reflects the tensions within the profession over the status to be accorded to recovered memories. As British researchers and clinicians involved in the debate, the authors of this article recognize the difficulties of reconciling clinical perspectives with the experimental approach of cognitive psychology. However, readers seeking guidance on theory and practice may be disappointed by the current report. Regarding theory, the authors maintain that more progress could have been achieved if the Working Group had shared a common vision of the way memory operates; it would have allowed more constructive dialogue regarding the significance of research on repression and traumatic memories. Regarding practice, there is no guidance as to how clinicians can amend their practice to avoid patient confabulation. The process of guideline development is described, together with some recent examples. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Decentralized nonlinear robust control of UAVs in close formation

This paper treats the design of a decentralized nonlinear robust control system for formation flying of multiple unmanned aerial vehicles (UAVs). In close formation, it is assumed that vortex of any UAV affects the motion of all the UAVs behind it. The forces produced by these vortices are complex functions of relative position co‐ordinates of the UAVs. In this paper, these forces are treated as unknown functions, which cannot be parameterized. Since the system is not invertible in the wind axes system, a simplified co‐ordinate system obtained from the wind axes system for which the velocity roll (bank angle) is zero, is considered for the design of the control system. A nonlinear robust control system for the separation trajectory control of the wing aircraft in the simplified wind coordinate system is derived. Uncertain functions and unmeasured variables are estimated using a high‐gain observer for the synthesis of the control system. Each wing UAV synthesizes its control law using its own state variables and the relative position of the preceding UAV with respect to the wing UAV. Thus the control system is decentralized since each UAV has to communicate (depending on sensors for position measurement) with at most one (preceding) UAV, and no data transmission from the remaining vehicles is required. Simulation results for two UAVs are presented which show precise separation trajectory control of each wing UAV in spite of the presence of unknown and unstructured vortex forces, while the lead aircraft maneuvers. Furthermore, these results confirm that when the wing aircraft is positioned properly in the vortex of the lead aircraft, it experiences reduction in its required flight power. Copyright © 2003 John Wiley & Sons, Ltd.     

Presence: Is it just pretending?

Our sense of presence in the real world helps regulate our behaviour within it by telling us about the status and effectiveness of our actions. As such, this ability offers us practical advantages in dealing effectively with the world. It is also an automatic or intuitive response to where and how we find ourselves in that it does not require conscious thought or deliberation. In contrast, the experience of presence or immersion in a movie, game or virtual environment is not automatic but is the product of our deliberate engagement with it, an engagement which first requires a disengagement or decoupling with the real world. Of course, we regularly decouple from the real world and embrace other, possible worlds every time we daydream, or engage in creative problem solving or, most importantly, for the purposes of this discussion, when we make-believe. We propose that make-believe is a plausible psychological mechanism which underpins the experience of mediated presence.     

Multi‐viewer autostereoscopic display with dynamically addressable holographic backlight

Abstract— The Multi‐User 3‐D Television Display (MUTED), designed to provide three‐dimensional television (3‐D TV) by the display of autostereoscopic imagery to multiple viewers, each of whom should enjoy freedom of movement, is described. Such an autostereoscopic display system, which allows multiple viewers simultaneously by the use of head tracking, was previously demonstrated for TV applications in the ATTEST project. However, the requirement for a dynamically addressable, steerable backlight presented several problems for the illumination source. The MUTED system demonstrates significant advances in the realization of a multi‐user autostereoscopic display, partly due to the provision of a dynamic backlight employing a novel holographic laser projector. Such a technology provides significant advantages in terms of brightness, efficiency, laser speckle, and the ability to correct for optical aberrations compared to both imaging and scanned‐beam projection technologies.     

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.     

A proteomic approach combining MS and bioinformatic analysis for the detection and identification of biomarkers of administration of exogenous human growth hormone in humans

An integrated MS-based proteomic approach is described that combines MALDI-MS and LC-MS with artificial neural networks for the identification of protein and peptide biomarkers associated with recombinant human growth hormone (rhGH) administration. Serum from exercised males administered with rhGH or placebo was analysed using ELISA to determine insulin-like growth factor-I concentrations. Diluted serum from rhGH- and placebo-treated subjects was analysed for protein biomarkers by MALDI-MS, whereas LC-MS was used to analyse tryptically digested ACN-depleted serum extracts for peptide biomarkers. Ion intensities and m/z values were used as inputs to artificial neural networks to classify samples into rhGH- and placebo-treated groups. Six protein ions (MALDI-MS) correctly classified 96% of samples into their respective groups, with a sensitivity of 91% (20 of 22 rhGH treated) and specificity of 100% (24 of 24 controls). Six peptide ions (LC-MS) were also identified and correctly classified 93% of samples with a sensitivity of 90% (19 of 21 rhGH treated) and a specificity of 95% (20 of 21 controls). The peptide biomarker ion with the highest significance was sequenced using LC-MS/MS and database searching and found to be associated with leucine-rich α-2-glycoprotein.     

A review of procedures to evolve quantum algorithms

There exist quantum algorithms that are more efficient than their classical counterparts; such algorithms were invented by Shor in 1994 and then Grover in 1996. A lack of invention since Grover’s algorithm has been commonly attributed to the non-intuitive nature of quantum algorithms to the classically trained person. Thus, the idea of using computers to automatically generate quantum algorithms based on an evolutionary model emerged. A limitation of this approach is that quantum computers do not yet exist and quantum simulation on a classical machine has an exponential order overhead. Nevertheless, early research into evolving quantum algorithms has shown promise. This paper provides an introduction into quantum and evolutionary algorithms for the computer scientist not familiar with these fields. The exciting field of using evolutionary algorithms to evolve quantum algorithms is then reviewed.

Nonlinear Adaptive Close Formation Control of Unmanned Aerial Vehicles

This paper treats the question of formationflight control of multiple unmanned aerial vehicles (UAVs). Inclose formation the wing UAV motion is affected by the vortexof the adjacent lead aircraft. The forces produced by these vorticesare complex functions of the relative position coordinates ofthe UAVs. In this paper, these forces are treated as unknownfunctions. For simplicity, it is assumed that the UAVs have autopilotsfor heading-, altitude-, and Mach-hold in the inner loops. Anadaptive control law is derived for the position control of thewing aircraft based on a backstepping design technique. In theclosed-loop system, commanded separation trajectories are asymptoticallytracked by each wing aircraft while the lead UAV is maneuvering.It is seen that an overparametrization in the design is essentialfor the decentralization of the control system. These resultsare applied to formation flight control of two UAVs and simulationresults are obtained. These results show that the wing UAV followsprecisely the reference separation trajectories in spite of theuncertainties in the aerodynamic coefficients, while the leadaircraft maneuvers.     

Experience with an efficient parallel kernel memory allocator

There has been great progress from the traditional allocation algorithms designed for small memories to more modern algorithms exemplified by McKusick's and Karels' allocator (McKusick MK, Karels MJ. Design of a general purpose memory allocator for the 4.3BSD UNIX kernel. In USENIX Conference Proceedings, Berkeley, CA, June 1988). Nonetheless, none of these algorithms have been designed to meet the needs of UNIX kernels supporting commercial data‐processing applications in a shared‐memory multiprocessor environment. On a shared‐memory multiprocessor, memory is a global resource. Therefore, allocator performance depends on synchronization primitives and manipulation of shared data as well as on raw CPU speed. Synchronization primitives and access to shared data depend on system bus interactions. The speed of system buses has not kept pace with that of CPUs, as witnessed by the ever‐larger caches found on recent systems. Thus, the performance of synchronization primitives and of memory allocators that use them have not received the full benefit of increased CPU performance. An earlier paper (McKenney PE, Slingwine J. Efficient kernel memory allocation on shared‐memory multiprocessors. In USENIX Conference Proceedings, Berkeley, CA, February 1993), describes an allocator designed to meet this situation. This article reviews the motivation for and design of the allocator and presents the experience gained during the seven years that the allocator has been in production use. Copyright © 2001 John Wiley & Sons, Ltd.