A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Filters
Assessment and Fragility of Byzantine Unreinforced Masonry Towers
2021
Infrastructures
Preserved Fulltext
. ≅ 0.15L v (1) Kouris and Karavezyroglou [17] proposed also an empirical relationship based on the ratio H/L and the modulus of elasticity of the soil Es: ≅ 0.153 ln(E ) + 0.2816( ) (2) Assuming standard ...
doi:10.3390/infrastructures6030040
fatcat:o3a3tos7kvayvo4z5qdbvf24q4
CascadeCNN: Pushing the performance limits of quantisation
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
This work presents CascadeCNN, an automated toolflow that pushes the quantisation limits of any given CNN model, to perform high-throughput inference by exploiting the computation time-accuracy trade-off. Without the need for retraining, a two-stage architecture tailored for any given FPGA device is generated, consisting of a low- and a high-precision unit. A confidence evaluation unit is employed between them to identify misclassified cases at run time and forward them to the high-precision
arXiv:1805.08743v1
fatcat:dyalnovqhbgehm2qrohh2nyxbi
more »
... t or terminate computation. Experiments demonstrate that CascadeCNN achieves a performance boost of up to 55% for VGG-16 and 48% for AlexNet over the baseline design for the same resource budget and accuracy.
Deploying Deep Neural Networks in the Embedded Space
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Recently, Deep Neural Networks (DNNs) have emerged as the dominant model across various AI applications. In the era of IoT and mobile systems, the efficient deployment of DNNs on embedded platforms is vital to enable the development of intelligent applications. This paper summarises our recent work on the optimised mapping of DNNs on embedded settings. By covering such diverse topics as DNN-to-accelerator toolflows, high-throughput cascaded classifiers and domain-specific model design, the
arXiv:1806.08616v1
fatcat:52xugpvnkzeuloufnsfp673oo4
more »
... nted set of works aim to enable the deployment of sophisticated deep learning models on cutting-edge mobile and embedded systems.
Stochastic Dynamic Analysis of Cultural Heritage Towers up to Collapse
2021
Buildings
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
This paper deals with the seismic vulnerability of monumental unreinforced masonry (URM) towers, the fragility of which has not yet been sufficiently studied. Thus, the present paper fills this gap by developing models to investigate the seismic response of URM towers up to collapse. On mount Athos, Greece, there exist more than a hundred medieval towers, having served mainly as campaniles or fortifications. Eight representative towers were selected for a thorough investigation to estimate
doi:10.3390/buildings11070296
fatcat:ylrspxe43zexzlua4ofyrpjuya
more »
... seismic response characteristics. Their history and architectural features are initially discussed and a two-step analysis follows: (i) limit analysis is performed to estimate the collapse mechanism and the locations of critical cracks, (ii) non-linear explicit dynamic analyses are then carried out, developing finite element (FE) simulations, with cracks modelled as interfacial surfaces to derive the capacity curves. A meaningful definition of the damage states is proposed based on the characteristics of their capacity curves, with the ultimate limit state related to collapse. The onset of slight damage-state is characterised by the formation and development of cracks responsible for the collapse mechanism of the structure. Apart from these two, another two additional limit states are also specified: the moderate damage-state and the extensive one. Fragility and vulnerability curves are finally generated which can help the assessment and preservation of cultural heritage URM towers.
VarSome: The Human Genomic Variant Search Engine
[article]
2018
bioRxiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Summary: VarSome.com is a search engine, aggregator and impact analysis tool for human genetic variation and a community-driven project aiming at sharing global expertise on human variants. Availability: VarSome is freely available at http://varsome.com.
doi:10.1101/367359
fatcat:x2rfqzidyvdetinqx6mms4zzxi
CascadeCNN: Pushing the Performance Limits of Quantisation in Convolutional Neural Networks
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
This work presents CascadeCNN, an automated toolflow that pushes the quantisation limits of any given CNN model, aiming to perform high-throughput inference. A two-stage architecture tailored for any given CNN-FPGA pair is generated, consisting of a low- and high-precision unit in a cascade. A confidence evaluation unit is employed to identify misclassified cases from the excessively low-precision unit and forward them to the high-precision unit for re-processing. Experiments demonstrate that
arXiv:1807.05053v1
fatcat:bn3sz2ewgjgfrgsmpzymnfmk5y
more »
... e proposed toolflow can achieve a performance boost up to 55% for VGG-16 and 48% for AlexNet over the baseline design for the same resource budget and accuracy, without the need of retraining the model or accessing the training data.
Toolflows for Mapping Convolutional Neural Networks on FPGAs: A Survey and Future Directions
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
In the past decade, Convolutional Neural Networks (CNNs) have demonstrated state-of-the-art performance in various Artificial Intelligence tasks. To accelerate the experimentation and development of CNNs, several software frameworks have been released, primarily targeting power-hungry CPUs and GPUs. In this context, reconfigurable hardware in the form of FPGAs constitutes a potential alternative platform that can be integrated in the existing deep learning ecosystem to provide a tunable balance
arXiv:1803.05900v1
fatcat:3gkwtxuahrghhmhz4nmkpqe7we
more »
... between performance, power consumption and programmability. In this paper, a survey of the existing CNN-to-FPGA toolflows is presented, comprising a comparative study of their key characteristics which include the supported applications, architectural choices, design space exploration methods and achieved performance. Moreover, major challenges and objectives introduced by the latest trends in CNN algorithmic research are identified and presented. Finally, a uniform evaluation methodology is proposed, aiming at the comprehensive, complete and in-depth evaluation of CNN-to-FPGA toolflows.
Approximate FPGA-Based LSTMs Under Computation Time Constraints
[chapter]
2018
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Recurrent Neural Networks and in particular Long Short-Term Memory (LSTM) networks have demonstrated state-of-the-art accuracy in several emerging Artificial Intelligence tasks. However, the models are becoming increasingly demanding in terms of computational and memory load. Emerging latency-sensitive applications including mobile robots and autonomous vehicles often operate under stringent computation time constraints. In this paper, we address the challenge of deploying computationally
doi:10.1007/978-3-319-78890-6_1
fatcat:p2q7b7snnnghrkfet43cqjlj6u
more »
... ing LSTMs at a constrained time budget by introducing an approximate computing scheme that combines iterative low-rank compression and pruning, along with a novel FPGA-based LSTM architecture. Combined in an end-to-end framework, the approximation method's parameters are optimised and the architecture is configured to address the problem of high-performance LSTM execution in time-constrained applications. Quantitative evaluation on a real-life image captioning application indicates that the proposed methods required up to 6.5× less time to achieve the same application-level accuracy compared to a baseline method, while achieving an average of 25× higher accuracy under the same computation time constraints.
A gradient elastic homogenisation model for brick masonry
2020
Engineering structures
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
CRediT authorship contribution statement Leonidas Alexandros S. Kouris: Conceptualization, Methodology, Software, Validation, Writing -original draft. Dionysios A. ...
Kouris, et al. ...
doi:10.1016/j.engstruct.2020.110311
fatcat:swfya47jjnc57ewpf4kjs7n5de
Detailed and simplified non-linear models for timber-framed masonry structures
2012
Journal of Cultural Heritage
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
This is the accepted version of the paper. This version of the publication may differ from the final published version. Permanent repository link: http://openaccess.city.ac.uk/3562/ Link to published version: http://dx. Abstract The need for improved methodologies to describe the post-elastic behaviour of existing structures in the framework of seismic vulnerability assessment has long been recognised. The study presented herein deals with the non-linear seismic response of timber-framed (T-F)
doi:10.1016/j.culher.2011.05.009
fatcat:iqmb7o5bkbfn3ek4jjunjaaiwe
more »
... asonry structures, such as those found in traditional edifices of cultural heritage. T-F masonry generally consists of masonry walls reinforced with timber elements, including horizontal and vertical elements, as well as X-type diagonal braces. Since the Bronze Age T-F buildings were common in regions where moderate to strong earthquakes were frequent. There is ample historical evidence that the embodiment of timber elements in masonry walls is closely related to earthquakes. The paper focuses on the description of the seismic response of T-F structures by means of a detailed analytical model. Although elastic analysis can adequately identify regions with high stresses, it fails to capture the redistribution of stresses and the ensuing failure mechanism. The simulation of T-F *Manuscript masonry is made here using a plasticity model. Nonlinear laws for the materials, such as a trilinear stress-strain curve for monotonic loading of timber and a Mohr-Coulomb contact law for wooden members, are used to express their behaviour under moderate and high stress levels. An associated flow rule is assumed and Hill's yield criterion is adopted with isotropic work-hardening. Masonry infills are not included in the model due to their insignificant contribution after the initial elastic stage of the response. The proposed finite element model is intended for a detailed non-linear static analysis of parts of a building. A simplified model using beam and link elements with non-linear axial springs is also developed, which is appropriate for 2-D non-linear analysis of common buildings. Both models are validated using experimental results of three T-F masonry walls obtained from the literature. Finally a non-linear static analysis of the façade of an existing building situated in the island of Lefkas, Greece is performed.
Familial Kaposi's Sarcoma: A Report of Five Cases from Greece
2014
Case Reports in Dermatological Medicine
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Introduction. Familial cases of Kaposi's sarcoma have rarely been reported. Kaposi's sarcoma is not uncommon in Greece; its incidence is estimated at 0.20 per 100.000 habitants, showing an increased predominance in the Peloponnese, in Southern Greece.Case Report. We describe five cases of familial clustering of KS originating from Greece.Discussion. The pathogenesis of familial Kaposi's sarcoma is still far from being completely understood. Genetic, environmental, and infectious factors have been incriminated.
doi:10.1155/2014/671631
pmid:25093124
pmcid:PMC4100369
fatcat:22f7n5g4pvdjzhgrbtubmjbdk4
A practice-oriented model for pushover analysis of a class of timber-framed masonry buildings
2014
Engineering structures
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Citation: Kouris, L.A.S. & Kappos, A. J. (2014). A practice-oriented model for pushover analysis of a class of timber-framed masonry buildings. ...
Kouris gratefully acknowledges the financial support provided by the Aristotle University of Thessaloniki for carrying out this work. ...
doi:10.1016/j.engstruct.2014.06.012
fatcat:6c6yc6sowrblffewf7sae4kdsi
Fragility Curves and Loss Estimation for Traditional Timber-Framed Masonry Buildings in Lefkas, Greece
[chapter]
2015
Computational Methods in Applied Sciences
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
Kouris gratefully acknowledges the financial support provided by the Aristotle University of Thessaloniki (Grant 16046/2011) for carrying out this work. ...
Details about the micro-model can be found in Kouris and Kappos (2012) . ...
This model has been based on the micro plasticity model of Kouris and Kappos (2012) . ...
doi:10.1007/978-3-319-16130-3_8
fatcat:ftirqmhmn5bujcbhwjaquuxso4
Dynamic Modification and Damage Propagation of a Two-Storey Full-Scale Masonry Building
2019
Advances in Civil Engineering
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
The progressive change of modal characteristics due to accumulated damage on an unreinforced masonry (URM) building is investigated. The stone URM building, submitted to five consecutive shakings, has been experimentally studied on the shaking table of EUCENTRE laboratory (Pavia, Italy). The dynamic characteristics of the test specimen are analytically estimated using frequency and state-space modal identification from ambient vibration stationary tests carried out before the strong motion
doi:10.1155/2019/2396452
fatcat:gu2ei44skfccppbwiiiiwyjzva
more »
... ient tests at various levels of damage. A singular value (SV) decomposition of the cross-correlation matrix of the acceleration response in the frequency domain is applied to determine the modal characteristics. In the time domain, the subspace state-space system identification is performed. Modal characteristics evolve from the initial linear state up to the ultimate collapse state in correlation with accumulated damage. Modal frequencies shorten with increasing intensity, whereas modal damping ratios are enhanced. Modal shapes also change with increasing level of accumulated damage. Comparing the evolution of modal characteristics, it is concluded that modal damping ratio shift can be better correlated with the system's actual performance giving a better representation of damage than that of natural frequency shift ratio or the modes difference.
Intelligent Identification of Structural Damage Based on the Curvature Mode and Wavelet Analysis Theory
2021
Advances in Civil Engineering
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Structural health monitoring is extensively used in new and old structures. During the process of monitoring, a large amount of data is generated. The selection of the appropriate methodology for the analysis of these data constitutes a major problem for maintenance personnel. Therefore, the purpose of this study is to programme data according to the curvature mode and wavelet transform theory to achieve automatic identification of structural damage and explore its applicability. First, the
doi:10.1155/2021/8847524
fatcat:mhpbgmxft5gzxoakekbt3zouda
more »
... ge model is established with finite element software, and the applicability of the theory is verified by analysing the wavelet coefficients before and after structural damage. Thereafter, a programme is written to achieve the automatic output of the structural damage location based on the finite element results and basic theory. Data fitting is then performed to estimate the degree of structural damage. To evaluate the effects of the practical application of the programme, experimental verification is conducted. The experimental results demonstrate that the programme can automatically output the damage location and avoid the occurrence of calculation errors.
« Previous
Showing results 1 — 15 out of 51 results