In this article, we review some of the classical methods used for quickly obtaining low-precision approximations to the elementary functions. Then, for each of the three main classes of elementary ...function algorithms (shift-and-add algorithms, polynomial or rational approximations, and table-based methods) and for the additional, specific to approximate computing, "bit-manipulation" techniques, we examine what can be done for obtaining very fast estimates of a function, at the cost of a (controlled) loss in terms of accuracy.
Abstract
Summary
Structural Variations (SV) are a major source of variability in the human genome that shaped its actual structure during evolution. Moreover, many human diseases are caused by SV, ...highlighting the need to accurately detect those genomic events but also to annotate them and assist their biological interpretation. Therefore, we developed AnnotSV that compiles functionally, regulatory and clinically relevant information and aims at providing annotations useful to (i) interpret SV potential pathogenicity and (ii) filter out SV potential false positive. In particular, AnnotSV reports heterozygous and homozygous counts of single nucleotide variations (SNVs) and small insertions/deletions called within each SV for the analyzed patients, this genomic information being extremely useful to support or question the existence of an SV. We also report the computed allelic frequency relative to overlapping variants from DGV (MacDonald et al., 2014), that is especially powerful to filter out common SV. To delineate the strength of AnnotSV, we annotated the 4751 SV from one sample of the 1000 Genomes Project, integrating the sample information of four million of SNV/indel, in less than 60 s.
Availability and implementation
AnnotSV is implemented in Tcl and runs in command line on all platforms. The source code is available under the GNU GPL license. Source code, README and Supplementary data are available at http://lbgi.fr/AnnotSV/.
Supplementary information
Supplementary data are available at Bioinformatics online.
Internet traffic classification has been the subject of intensive study since the birth of the Internet itself. Indeed, the evolution of approaches for traffic classification can be associated with ...the evolution of the Internet itself and with the adoption of new services and the emergence of novel applications and communication paradigms. Throughout the years many approaches have been proposed for addressing technical issues imposed by such novel services. Deep-Packet Inspection (DPI) has been a very important research topic within the traffic classification field and its concept consists of the analysis of the contents of the captured packets in order to accurately and timely discriminate the traffic generated by different Internet protocols. DPI was devised as a means to address several issues associated with port-based and statistical-based classification approaches in order to achieve an accurate and timely traffic classification. Many research works proposed different DPI schemes while many open-source modules have also become available for deployment. Surveys become then valuable tools for performing an overall analysis, study and comparison between the several proposed methods. In this paper we present a survey in which a complete and thorough analysis of the most important open-source DPI modules is performed. Such analysis comprises an evaluation of the classification accuracy, through a common set of traffic traces with ground truth, and of the computational requirements. In this manner, this survey presents a technical assessment of DPI modules and the analysis of the obtained evaluation results enable the proposal of general guidelines for the design and implementation of more adequate DPI modules.
Algorithms for Triple-Word Arithmetic Fabiano, Nicolas; Muller, Jean-Michel; Picot, Joris
I.E.E.E. transactions on computers/IEEE transactions on computers,
2019-Nov.-1, 2019-11-1, 2019-11, Letnik:
68, Številka:
11
Journal Article
Recenzirano
Odprti dostop
Triple-word arithmetic consists in representing high-precision numbers as the unevaluated sum of three floating-point numbers (with "nonoverlapping" constraints that are explicited in the paper). We ...introduce and analyze various algorithms for manipulating triple-word numbers: rounding a triple-word number to a floating-point number, adding, multiplying, dividing, and computing square-roots of triple-word numbers, etc. We compare our algorithms, implemented in the Campary library, with other solutions of comparable accuracy. It turns out that our new algorithms are significantly faster than what one would obtain by just using the usual floating-point expansion algorithms in the special case of expansions of length 3.
This installment of Computer’s series highlighting the work published in IEEE Computer Society journals comes from IEEE Transactions on Emerging Topics in Computing.
Aim
This study compared neurodevelopmental screening questionnaires completed when preterm‐born children reached 2 years of corrected age with social communication skills at 5.5 years of age.
Methods
...Eligible subjects were born in 2011 at 24–34 weeks of gestation, participated in a French population‐based epidemiological study and were free of motor and sensory impairment at 2 years of corrected age. The Ages and Stages Questionnaire (ASQ) and the Modified Checklist for Autism in Toddlers (M‐CHAT) were used at 2 years and the Social Communication Questionnaire (SCQ) at 5.5 years of age.
Results
We focused on 2119 children. At 2 years of corrected age, the M‐CHAT showed autistic traits in 20.7%, 18.5% and 18.2% of the children born at 24–26, 27–31 and 32–34 weeks of gestation, respectively (p = 0.7). At 5.5 years of age, 12.6%, 12.7% and 9.6% risked social communication difficulties, with an SCQ score ≥90th percentile (p = 0.2). A positive M‐CHAT score at 2 years was associated with higher risks of social communication difficulties at 5.5 years of age (odds ratio 3.46, 95% confidence interval 2.04–5.86, p < 0.001). Stratifying ASQ scores produced similar results.
Conclusion
Using parental neurodevelopmental screening questionnaires for preterm‐born children helped to identify the risk of later social communication difficulties.
Assume we use a binary floating-point arithmetic and that <inline-formula><tex-math notation="LaTeX">\operatorname{RN}</tex-math> <mml:math><mml:mo form="prefix">RN</mml:mo></mml:math><inline-graphic ...xlink:href="muller-ieq1-3294986.gif"/> </inline-formula> is the round-to-nearest function. Also assume that <inline-formula><tex-math notation="LaTeX">c</tex-math> <mml:math><mml:mi>c</mml:mi></mml:math><inline-graphic xlink:href="muller-ieq2-3294986.gif"/> </inline-formula> is a constant or a real function of one or more variables, and that we have at our disposal a correctly rounded implementation of <inline-formula><tex-math notation="LaTeX">c</tex-math> <mml:math><mml:mi>c</mml:mi></mml:math><inline-graphic xlink:href="muller-ieq3-3294986.gif"/> </inline-formula>, say <inline-formula><tex-math notation="LaTeX">\hat{c}= \operatorname{RN}(c)</tex-math> <mml:math><mml:mrow><mml:mover accent="true"><mml:mi>c</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mo>=</mml:mo><mml:mo form="prefix">RN</mml:mo><mml:mrow><mml:mo>(</mml:mo><mml:mi>c</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq4-3294986.gif"/> </inline-formula>. For evaluating <inline-formula><tex-math notation="LaTeX">x \cdot c</tex-math> <mml:math><mml:mrow><mml:mi>x</mml:mi><mml:mo>·</mml:mo><mml:mi>c</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq5-3294986.gif"/> </inline-formula> (resp. <inline-formula><tex-math notation="LaTeX"> x / c</tex-math> <mml:math><mml:mrow><mml:mi>x</mml:mi><mml:mo>/</mml:mo><mml:mi>c</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq6-3294986.gif"/> </inline-formula> or <inline-formula><tex-math notation="LaTeX">c / x</tex-math> <mml:math><mml:mrow><mml:mi>c</mml:mi><mml:mo>/</mml:mo><mml:mi>x</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq7-3294986.gif"/> </inline-formula>), the natural way is to replace it by <inline-formula><tex-math notation="LaTeX">\operatorname{RN}(x \cdot \hat{c})</tex-math> <mml:math><mml:mrow><mml:mo form="prefix">RN</mml:mo><mml:mo>(</mml:mo><mml:mi>x</mml:mi><mml:mo>·</mml:mo><mml:mover accent="true"><mml:mi>c</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mo>)</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq8-3294986.gif"/> </inline-formula> (resp. <inline-formula><tex-math notation="LaTeX"> \operatorname{RN}(x / \hat{c})</tex-math> <mml:math><mml:mrow><mml:mo form="prefix">RN</mml:mo><mml:mo>(</mml:mo><mml:mi>x</mml:mi><mml:mo>/</mml:mo><mml:mover accent="true"><mml:mi>c</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mo>)</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq9-3294986.gif"/> </inline-formula> or <inline-formula><tex-math notation="LaTeX">\operatorname{RN}(\hat{c}/ x)</tex-math> <mml:math><mml:mrow><mml:mo form="prefix">RN</mml:mo><mml:mo>(</mml:mo><mml:mover accent="true"><mml:mi>c</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mo>/</mml:mo><mml:mi>x</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq10-3294986.gif"/> </inline-formula>), that is, to call function <inline-formula><tex-math notation="LaTeX">\hat{c}</tex-math> <mml:math><mml:mover accent="true"><mml:mi>c</mml:mi><mml:mo>^</mml:mo></mml:mover></mml:math><inline-graphic xlink:href="muller-ieq11-3294986.gif"/> </inline-formula> and to perform a floating-point multiplication or division. This can be generalized to the approximation of <inline-formula><tex-math notation="LaTeX">n/d</tex-math> <mml:math><mml:mrow><mml:mi>n</mml:mi><mml:mo>/</mml:mo><mml:mi>d</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq12-3294986.gif"/> </inline-formula> by <inline-formula><tex-math notation="LaTeX">\operatorname{RN}(\hat{n}/\hat{d})</tex-math> <mml:math><mml:mrow><mml:mo form="prefix">RN</mml:mo><mml:mo>(</mml:mo><mml:mover accent="true"><mml:mi>n</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mo>/</mml:mo><mml:mover accent="true"><mml:mi>d</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mo>)</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq13-3294986.gif"/> </inline-formula> and the approximation of <inline-formula><tex-math notation="LaTeX">n \cdot d</tex-math> <mml:math><mml:mrow><mml:mi>n</mml:mi><mml:mo>·</mml:mo><mml:mi>d</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq14-3294986.gif"/> </inline-formula> by <inline-formula><tex-math notation="LaTeX">\operatorname{RN}(\hat{n} \cdot \hat{d})</tex-math> <mml:math><mml:mrow><mml:mo form="prefix">RN</mml:mo><mml:mo>(</mml:mo><mml:mover accent="true"><mml:mi>n</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mo>·</mml:mo><mml:mover accent="true"><mml:mi>d</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mo>)</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq15-3294986.gif"/> </inline-formula>, where <inline-formula><tex-math notation="LaTeX">\hat{n} = \operatorname{RN}(n)</tex-math> <mml:math><mml:mrow><mml:mover accent="true"><mml:mi>n</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mo>=</mml:mo><mml:mo form="prefix">RN</mml:mo><mml:mrow><mml:mo>(</mml:mo><mml:mi>n</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq16-3294986.gif"/> </inline-formula> and <inline-formula><tex-math notation="LaTeX">\hat{d} = \operatorname{RN}(d)</tex-math> <mml:math><mml:mrow><mml:mover accent="true"><mml:mi>d</mml:mi><mml:mo>^</mml:mo></mml:mover><mml:mo>=</mml:mo><mml:mo form="prefix">RN</mml:mo><mml:mrow><mml:mo>(</mml:mo><mml:mi>d</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq17-3294986.gif"/> </inline-formula>, and <inline-formula><tex-math notation="LaTeX">n</tex-math> <mml:math><mml:mi>n</mml:mi></mml:math><inline-graphic xlink:href="muller-ieq18-3294986.gif"/> </inline-formula> and <inline-formula><tex-math notation="LaTeX">d</tex-math> <mml:math><mml:mi>d</mml:mi></mml:math><inline-graphic xlink:href="muller-ieq19-3294986.gif"/> </inline-formula> are functions for which we have at our disposal a correctly rounded implementation. We discuss tight error bounds in ulps of such approximations. From our results, one immediately obtains tight error bounds for calculations such as <inline-formula><tex-math notation="LaTeX">\mathtt {x * pi}</tex-math> <mml:math><mml:mrow><mml:mi mathvariant="monospace">x</mml:mi><mml:mo>*</mml:mo><mml:mi mathvariant="monospace">pi</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq20-3294986.gif"/> </inline-formula>, <inline-formula><tex-math notation="LaTeX">\mathtt {ln(2)/x}</tex-math> <mml:math><mml:mrow><mml:mi mathvariant="monospace">ln</mml:mi><mml:mo>(</mml:mo><mml:mn mathvariant="monospace">2</mml:mn><mml:mo>)</mml:mo><mml:mo>/</mml:mo><mml:mi mathvariant="monospace">x</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq21-3294986.gif"/> </inline-formula>, <inline-formula><tex-math notation="LaTeX">\mathtt {x/(y+z)}</tex-math> <mml:math><mml:mrow><mml:mi mathvariant="monospace">x</mml:mi><mml:mo>/</mml:mo><mml:mo>(</mml:mo><mml:mi mathvariant="monospace">y</mml:mi><mml:mo>+</mml:mo><mml:mi mathvariant="monospace">z</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq22-3294986.gif"/> </inline-formula>, <inline-formula><tex-math notation="LaTeX">\mathtt {(x+y)*z}</tex-math> <mml:math><mml:mrow><mml:mo>(</mml:mo><mml:mi mathvariant="monospace">x</mml:mi><mml:mo>+</mml:mo><mml:mi mathvariant="monospace">y</mml:mi><mml:mo>)</mml:mo><mml:mo>*</mml:mo><mml:mi mathvariant="monospace">z</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq23-3294986.gif"/> </inline-formula>, <inline-formula><tex-math notation="LaTeX">\mathtt {x/sqrt(y)}</tex-math> <mml:math><mml:mrow><mml:mi mathvariant="monospace">x</mml:mi><mml:mo>/</mml:mo><mml:mi mathvariant="monospace">sqrt</mml:mi><mml:mo>(</mml:mo><mml:mi mathvariant="monospace">y</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq24-3294986.gif"/> </inline-formula>, <inline-formula><tex-math notation="LaTeX">\mathtt {sqrt(x)/{y}}</tex-math> <mml:math><mml:mrow><mml:mi mathvariant="monospace">sqrt</mml:mi><mml:mo>(</mml:mo><mml:mi mathvariant="monospace">x</mml:mi><mml:mo>)</mml:mo><mml:mo>/</mml:mo><mml:mi mathvariant="monospace">y</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq25-3294986.gif"/> </inline-formula>, <inline-formula><tex-math notation="LaTeX">\mathtt {(x+y)(z+t)}</tex-math> <mml:math><mml:mrow><mml:mo>(</mml:mo><mml:mi mathvariant="monospace">x</mml:mi><mml:mo>+</mml:mo><mml:mi mathvariant="monospace">y</mml:mi><mml:mo>)</mml:mo><mml:mo>(</mml:mo><mml:mi mathvariant="monospace">z</mml:mi><mml:mo>+</mml:mo><mml:mi mathvariant="monospace">t</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq26-3294986.gif"/> </inline-formula>, <inline-formula><tex-math notation="LaTeX">\mathtt {(x+y)/(z+t)}</tex-math> <mml:math><mml:mrow><mml:mo>(</mml:mo><mml:mi mathvariant="monospace">x</mml:mi><mml:mo>+</mml:mo><mml:mi mathvariant="monospace">y</mml:mi><mml:mo>)</mml:mo><mml:mo>/</mml:mo><mml:mo>(</mml:mo><mml:mi mathvariant="monospace">z</mml:mi><mml:mo>+</mml:mo><mml:mi mathvariant="monospace">t</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq27-3294986.gif"/> </inline-formula>, <inline-formula><tex-math notation="LaTeX">\mathtt {(x+y)/(zt)}</tex-math> <mml:math><mml:mrow><mml:mo>(</mml:mo><mml:mi mathvariant="monospace">x</mml:mi><mml:mo>+</mml:mo><mml:mi mathvariant="monospace">y</mml:mi><mml:mo>)</mml:mo><mml:mo>/</mml:mo><mml:mo>(</mml:mo><mml:mi mathvariant="monospace">zt</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="muller-ieq28-3294986.gif"/> </inline-formula>, etc. in floating-point arithmetic.
In recent years, there has been a wide array of research studies published on parental mental health and stress following very preterm birth. This review aims at reviewing the prevalence and risk ...factors of long-term parental depression, anxiety, post-traumatic stress symptoms and parenting stress following very preterm birth.
We searched PubMed, PsychINFO and Web of Science for descriptive, cross-sectional and longitudinal studies published between January 2013 and August 2022.
45 studies met our inclusion criteria. In the first two years, depression, anxiety, post-traumatic stress symptoms and parenting stress were present in ∼20 % of mothers of extreme and very low birth weight (E/VLBW) infants. Long-term psychological distress symptoms could be observed, although few studies have focused on symptoms into school age and longer. Fathers of VLBW infants might experience more psychological distress as well, however, they were only included in ten studies. We found that parental distress is more common when the co-parent is struggling with mental health symptoms. Many risk factors were identified such as social risk, history of mental illness, interpersonal factors (i.e. social support) and child-related factors (i.e. intraventricular hemorrhage, disability, use of medical equipment at home).
Several studies have methodological issues, such as a lack of control of known confounders and there is a large variety of measures employed.
Important risk factors for stress and mental health symptoms were identified. More evidence is needed to determine if long-term symptoms persist into school age. Research should focus on taking a family-based approach in order to identify preventive strategies and resilience factors in parents of VLBW infants.
•Mental health and parenting stress symptomatology is common in the two first years following very preterm birth.•Long term outcomes on mental health could be related to personal and interpersonal risk factors.•Fathers have been less considered in research on mental health and stress outcomes.
PDF
