78 resultados para invention and memory
Resumo:
A novel design for multibit convolver circuits is described. The circuits take the form of systolic arrays of simple one-bit processor and memory cells, with the result that they can operate at very high data rates and should be easy to implement using VLSI technology. An efficient method for handling two's complement data within the array is described and the relative advantages of this convolver design compared with more conventional circuits is discussed.
Resumo:
We propose a dynamic verification approach for large-scale message passing programs to locate correctness bugs caused by unforeseen nondeterministic interactions. This approach hinges on an efficient protocol to track the causality between nondeterministic message receive operations and potentially matching send operations. We show that causality tracking protocols that rely solely on logical clocks fail to capture all nuances of MPI program behavior, including the variety of ways in which nonblocking calls can complete. Our approach is hinged on formally defining the matches-before relation underlying the MPI standard, and devising lazy update logical clock based algorithms that can correctly discover all potential outcomes of nondeterministic receives in practice. can achieve the same coverage as a vector clock based algorithm while maintaining good scalability. LLCP allows us to analyze realistic MPI programs involving a thousand MPI processes, incurring only modest overheads in terms of communication bandwidth, latency, and memory consumption. © 2011 IEEE.
Resumo:
There is considerable interest in creating embedded, speech recognition hardware using the weighted finite state transducer (WFST) technique but there are performance and memory usage challenges. Two system optimization techniques are presented to address this; one approach improves token propagation by removing the WFST epsilon input arcs; another one-pass, adaptive pruning algorithm gives a dramatic reduction in active nodes to be computed. Results for memory and bandwidth are given for a 5,000 word vocabulary giving a better practical performance than conventional WFST; this is then exploited in an adaptive pruning algorithm that reduces the active nodes from 30,000 down to 4,000 with only a 2 percent sacrifice in speech recognition accuracy; these optimizations lead to a more simplified design with deterministic performance.
Resumo:
FastFlow is a programming framework specifically targeting cache-coherent shared-memory multi-cores. It is implemented as a stack of C++ template libraries built on top of lock-free (and memory fence free) synchronization mechanisms. Its philosophy is to combine programmability with performance. In this paper a new FastFlow programming methodology aimed at supporting parallelization of existing sequential code via offloading onto a dynamically created software accelerator is presented. The new methodology has been validated using a set of simple micro-benchmarks and some real applications. © 2011 Springer-Verlag.
Resumo:
Purpose: To compare long-term cognitive outcomes of patients treated with surgical clipping or endovascular coiling after subarachnoid haemorrhage (SAH). Method: Retrospective matched cohort study assessed neuropsychological functioning at least 12 months after aneurysmal SAH treatment. Fourteen patients treated by endovascular coiling and nine patients treated by surgical clipping participated. After gaining written consent, a comprehensive neuropsychological battery was completed. Standardised tests were employed to assess pre-morbid and current intellectual functioning (IQ), attention, speed of information processing, memory and executive function as well as psychosocial functioning and affect. Results: Treatment groups were not significantly different in terms of age, pre-morbid IQ, time from injury to treatment or time since injury. A significant effect of treatment on full-scale IQ score (p = 0.025), performance IQ (p = 0.045) and verbal IQ score (p = 0.029), all favouring the coiled group was observed. A medium effect size between groups difference in immediate memory (p = 0.19, partial ?(2) = 0.08) was also observed. No significant between group differences on attention, executive functioning and speed of information processing measures or mood and psychosocial functioning were noted. Both groups reported increased anxiety and memory, attention and speed of information processing deficits relative to normative data. Conclusions: Study findings indicate fewer cognitive deficits following endovascular coiling. Cognitive deficits in the clipped group may be due in part to the invasive nature of neurosurgical clipping. Further prospective research with regard to long-term cognitive and emotional outcomes is warranted. [Box: see text].
Resumo:
Objectives. The hypotheses that automatic, non-volitional, attentional and memory biases for addiction-related constructs exist is tested with compulsive gamblers.
Design. An independent groups design was employed. Processing of gambling, compared to neutral and drug-related information was examined in 15 gamblers recruited from new members of Gamblers Anonymous. Comparisons were made with the performance of their spouses (N = 15) to help distinguish addiction mechanisms from more non-specific emotional experiences with gambling, and an independent control group (N = 15), recruited from the staff and students of a university department.
Methods. A modified Stroop procedure was first employed. Automative cognitive interference was assessed relatively, by comparing colour-naming times on the gambling, drug and neutral Stroops. A subsequent word-stem completion task of implicit memory was then used to assess selective and automatic priming of the gambling constructs in memory.
Results. Only the gamblers showed selective and automatic interference for gambling-related constructs on the Stroop task. Spouses behaved like the control group on this task. An implicit memory bias for gambling-related words was statistically detected only in the gamblers compared to the control group, although the trend was similar in the comparison with spouses. Further evidence for the specificity of these effects was obtained in subgroup comparisons involving fruit-machine with racing gamblers.
Conclusions. Results are generally consistent with an automaticity in the cognitive biases gamblers show for gambling-related information. implications for cognitive understanding and treatments are highlighted.
Resumo:
“Sounds of the City” is a large-scale community project and exhibition commissioned by the Metropolitan Arts Centre (MAC) and led by artists from the Sonic Arts Research Centre (SARC), Queen’s University Belfast. Over a four-month period, the artists worked together with two intergenerational groups in Belfast with the aim of addressing specific sound qualities of places, events and stories. Themes that surfaced from this process constitute the basis for the exhibition which promotes listening as a form of intersecting daily life, identity and memory. Five installations address aural contexts ranging from Belfast’s industrial heritage to the local family home. These are shaped by present and past experiences of workshop participants at Dee Street Community Centre in East Belfast and Tar Isteach in North Belfast. The themes and contents of these installations center upon the relationship between sound and memory, sound and place, and the documentation of everyday personal auditory experience.
All materials exhibited have emerged through workshops, interviews and field-recording sessions. Workshops acted as a basis from which to inform each group about the project’s aims, methods of listening, methods of documenting sound and the wider areas of soundscape studies and acoustic ecology. They also provided a central point allowing participants to organize outside activities and share material for exhibition.
“Sounds of the City” explores the relationship between sound and community through everyday life and presents a dynamic and ever-changing soundscape that shapes Belfast’s identity.
Sounds of the City has been exhibited at the MAC, Belfast 2012 and at Espaço Ecco, Brasilia 2013.
Resumo:
We study the interplay between forgetful and memory-keeping evolution enforced on a two-level system by a multi-spin environment whose elements are coupled to local bosonic baths. Contrarily to the expectation that any non-Markovian effect would be buried by the forgetful mechanism induced by the spin-bath coupling, one can actually induce a full Markovian-to-non-Markovian transition of the two-level system's dynamics, controllable by parameters such as the mismatch between the energy of the two-level system and of the spin environment. For a symmetric coupling, the amount of non-Markovianity surprisingly grows with the number of decoherence channels. DOI: 10.1103/PhysRevA.87.022317
Resumo:
Polish Academy of Warsaw - War and Memory conference
September 2012
Resumo:
Objectives: Multitasking is a challenging aspect of human behavior, especially if the concurrently performed tasks are different in nature. Several studies demonstrated pronounced performance decrements (dual-task costs) in older adults for combinations of cognitive and motor tasks. However, patterns of costs among component tasks differed across studies and reasons for participants' resource allocation strategies remained elusive.
Method:We investigated young and older adults' multitasking of a working memory task and two sensorimotor tasks, one with low (finger force control) and one with high ecological relevance (postural control). The tasks were performed in single-, dual-, and triple-task contexts.
Results: Working memory accuracy was reduced in dual-task contexts with either sensorimotor task and deteriorated further under triple-task conditions. Postural and force performance deteriorated with age and task difficulty in dual-task contexts. However, in the triple-task context with its maximum resource demands, older adults prioritized postural control
over both force control and memory.
Discussion: Our results identify ecological relevance as the key factor in older adults’ multitasking.
Resumo:
The hundredth anniversary of the outbreak of the First World War is only the first of a large number of major European historical anniversaries that will occur in the coming four years. Other twentieth-century anniversaries include that of the Russian Revolution and the Easter Uprising; notable corollaries from earlier centuries include the Battle of Bannockburn, the Hanoverian succession, the Battle of Waterloo and, perhaps most significant of all, the five hundredth anniversary of the Lutheran Reformation. Rather than commission special issues or other features to tie in to individual anniversaries centred on or relevant to German history in a manner which repeats unthinkingly the conventions of scholarly and popular culture, the editors elected to reflect more fundamentally on what might be at stake in major anniversaries for professional scholars of history. In anticipation of the major wave of scholarly and popular publications, commemorative activities and memory conflicts that each of these will generate, and in order to reflect upon the dynamics of German history, memory and commemoration in a more overtly comparative context, the editors invited a number of scholars working on different national histories to reflect on the possibilities and potential pitfalls such anniversaries offer to historians who tie their work in to such moments. They are Jörg Arnold (Nottingham), Thomas A. Brady (Berkeley), Fearghal McGarry (Queen’s University, Belfast), Tim Grady (Chester) and Dan Healey (St Antony’s College, Oxford). The questions were posed by the editors.
Resumo:
Schools of nursing continuously strive to facilitate learning through student engagement and teaching strategies that encourage active learning. This paper reports on the successful use of mind mapping, an underutilised and underdeveloped strategy, to enhance teaching and learning in undergraduate nurse education (Spencer et al., 2013). Mind mapping or concept mapping has been defined in the literature as a visual representation of one’s thoughts and ideas (Abel and Freeze, 2006). It is characterised by colour, images and text in a graphical, nonlinear style. Mind maps promote the linking of concepts and capitalise on the brain’s natural aptitude for visual recognition to enhance learning and memory recall (Buzan, 2006). Traditional teaching strategies depend on linear processes, which in comparison lack engagement, associations and creativity (Spencer et al., 2013). Mind mapping was introduced to nursing students undertaking modules in ‘Dimensions of Care’ and ‘Care Delivery’ on year two of the nursing degree programme in Queen’s University Belfast. The aim of introducing mind mapping was to help students make the critical link between the pathophysiology of conditions studied and the provision of informed, safe and effective patient care, which had challenged previous student cohorts. Initially maps were instructor-made as described by Boley (2008), as a template for note taking during class and as a study aid. However, students rapidly embraced the strategy and started creating their own mind maps. Meaningful learning occurs when students engage with concepts and organise them independently in a way significant to them (Buzan, 2006). Students reported high levels of satisfaction to this teaching approach. This paper will present examples of the mind maps produced and explore how mind mapping can be further utilised within the undergraduate nursing curriculum.
Resumo:
Multi-bit trie is a popular approach performing the longest prefix matching for packet classification. However, it requires a long lookup time and inefficiently consumes memory space. This paper presents an in-depth study of different variations of multi-bit trie for IP address lookup. Our main aim is to study a method of data structure which reduces memory space. The proposed approach has been implemented using the label method in two approaches. Both methods present better results regarding lookup speed, update time and memory bit consumptions.
Resumo:
This paper addresses the problem of learning Bayesian network structures from data based on score functions that are decomposable. It describes properties that strongly reduce the time and memory costs of many known methods without losing global optimality guarantees. These properties are derived for different score criteria such as Minimum Description Length (or Bayesian Information Criterion), Akaike Information Criterion and Bayesian Dirichlet Criterion. Then a branch-and-bound algorithm is presented that integrates structural constraints with data in a way to guarantee global optimality. As an example, structural constraints are used to map the problem of structure learning in Dynamic Bayesian networks into a corresponding augmented Bayesian network. Finally, we show empirically the benefits of using the properties with state-of-the-art methods and with the new algorithm, which is able to handle larger data sets than before.