5916 research outputs found
Sort by
Pytorch and cedr: Enabling deployment of machine learning models on heterogeneous computing systems
No full textThe PyTorch programming interface enables efficient deployment of machine learning models, leveraging the parallelism offered by GPU architectures. In this study, we present the integration of the PyTorch framework with a compiler and runtime ecosystem. Our aim is to demonstrate the ability to deploy PyTorch-based models on FPGA-based SoC platforms, without requiring users to possess prior FPGA-based design experience. The proposed PyTorch model transformation approach expands the range of hardware architectures that PyTorch developers can target, enabling them to take advantage of the energy-efficient execution provided by heterogeneous computing systems. Our experiments involve compiling and executing real-life applications on heterogeneous SoC configurations emulated on the Xilinx Zynq Ultrascale+ ZCU102 system. We showcase our ability to deploy three distinct PyTorch applications, encompassing object detection, visual geometry group (VGG), and speech classification, using the integrated compiler and runtime system without loss of model accuracy. Furthermore, we extend our analysis by evaluating dynamically arriving workload scenarios, consisting of a mix of PyTorch models and non-PyTorch-based applications. Through these experiments, we vary the hardware composition and scheduling heuristics. Our findings indicate that when PyTorch-based applications coexist with unrelated applications, our integrated scheduler fairly dispatches tasks to the FPGA platform's accelerator and CPU cores, without compromising the target throughput for each application.United States Department of Defense US Air Force Research Laboratory ; United States Department of Defense Defense Advanced Research Projects Agency (DARPA
Screening for eligibility: access and resistance in Istanbul’s food banks
No full textIntroduced in the 2000s as a component of social welfare reforms, the means test determines the eligibility of aid applicants based on previously set income categories. Replacing local committees that decided eligibility, this centralized and digitalized screening process rests on information infrastructures that are mostly invisible. This paper argues that the ways in which applicants contest the outcome of the means test, subvert the eligibility requirements, and go around the screening processes, make visible these otherwise-mostly invisible information infrastructures. Through a discussion of the contestations, subversions, and go-arounds applicants use (not always successfully) to receive emergency food relief from municipal food banks in Istanbul, the paper shows that these information infrastructures not only appear as if they are value-neutral and apolitical, but in so doing, they also serve as useful tools for obscuring who the actual decision makers are
Probabilistic seismic microzonation for ground shaking intensity, a case study in Türkiye
No full textThe purpose of seismic microzonation is to estimate earthquake characteristics on the ground surface based on a probabilistic approach to mitigate earthquake damage in the foreseeable future for the new buildings, as well as for the existing building stock. The probabilistic analysis and related results are very important from an engineering perspective since the nature of the problem can only be dealt with in a probabilistic manner. The uncertainties associated with these analyses may be large due to the uncertainties in source characteristics, soil profile, soil properties, and building inventory. At this stage, the probability distribution of the related earthquake parameters on the ground surface may be determined based on hazard-compatible input acceleration-time histories, site profiles, and dynamic soil properties. One option, the variability in earthquake source and path effects may be considered using a large number of acceleration records compatible with the site-dependent earthquake hazard. Likewise, large numbers of soil profiles may be used to account for the site-condition variability. The seismic microzonation methodology is proposed based on the probabilistic assessment of these factors involved in site response analysis. The second important issue in seismic microzonation procedure is the selection of microzonation parameters. The purpose being mitigation of structural damage, it is possible to adopt earthquake parameters like cumulative average velocity (CAV) or Housner intensity (HI) that was observed to have better correlation with building damage after earthquakes. A seismic microzonation procedure will be developed with respect to ground shaking intensity considering probabilistic values of the cumulative average velocity (CAV) or Housner intensity (HI)
Azimuthal anisotropy of dijet events in PbPb collisions at √sNN=5.02 TeV
No full textThe path-length dependent parton energy loss within the dense partonic medium created in lead-lead collisions at a nucleon-nucleon center-of-mass energy of root s(NN) = 5.02TeV is studied by determining the azimuthal anisotropies for dijets with high transverse momentum. The data were collected by the CMS experiment in 2018 and correspond to an integrated luminosity of 1.69 nb(-1). For events containing back-to-back jets, correlations in relative azimuthal angle and pseudorapidity (eta) between jets and hadrons, and between two hadrons, are constructed. The anisotropies are expressed as the Fourier expansion coefficients v(n), n = 2-4 of these azimuthal distributions. The dijet v(n) values are extracted from long-range (1.5 < vertical bar Delta eta vertical bar < 2.5) components of these correlations, which suppresses the background contributions from jet fragmentation processes. Positive dijet v(2) values are observed which increase from central to more peripheral events, while the v(3) and v(4) values are consistent with zero within experimental uncertainties.BMBWF and FWF (Austria); FNRS and FWO (Belgium); CNPq, CAPES, FAPERJ, FAPERGS, and FAPESP (Brazil); MES and BNSF (Bulgaria); CERN; CAS, MoST, and NSFC (China); MINCIENCIAS (Colombia); MSES and CSF (Croatia); RIF (Cyprus); SENESCYT (Ecuador); MoER, ERC PUT and ERDF (Estonia); Academy of Finland, MEC, and HIP (Finland); CEA and CNRS/IN2P3 (France); BMBF, DFG, and HGF (Germany); GSRI (Greece); NKFIH (Hungary); DAE and DST (India); IPM (Iran); SFI (Ireland); INFN (Italy); MSIP and NRF (Republic of Korea); MES (Latvia); LAS (Lithuania); MOE and UM (Malaysia); BUAP, CINVESTAV, CONACYT, LNS, SEP, and UASLP-FAI (Mexico); MOS (Montenegro); MBIE (New Zealand); PAEC (Pakistan); MES and NSC (Poland); FCT (Portugal); MESTD (Serbia); MCIN/AEI and PCTI (Spain); MOSTR (Sri Lanka); Swiss Funding Agencies (Switzerland); MST (Taipei); MHESI and NSTDA (Thailand); TUBITAK and TENMAK (Turkey); NASU (Ukraine); STFC (United Kingdom); DOE and NSF (USA).
Individuals have received support from the Marie-Curie program and the European Research Council and Horizon 2020 Grant, contract Nos. 675440, 724704, 752730, 758316, 765710, 824093, 884104, and COST Action CA16108 (European Union); the Leventis Foundation; the Alfred P. Sloan Foundation; the Alexander von Humboldt Foundation; the Belgian Federal Science Policy Office; the Fonds pour la Formation a la Recherche dans l'Industrie et dans l'Agriculture (FRIA-Belgium); the Agentschap voor Innovatie door Wetenschap en Technologie (IWT-Belgium); the F.R.S.-FNRS and FWO (Belgium) under the "Excellence of Science -EOS" -be.h project n. 30820817; the Beijing Municipal Science & Technology Commission, No. Z191100007219010; the Ministry of Education, Youth and Sports (MEYS) of the Czech Republic; the Hellenic Foundation for Research and Innovation (HFRI), Project Number 2288 (Greece); the Deutsche Forschungsgemeinschaft (DFG), under Germany's Excellence Strategy -EXC 2121 "Quantum Universe" -390833306, and under project number 400140256 -GRK2497; the Hungarian Academy of Sciences, the New National Excellence Program -UNKP, the NKFIH research grants K 124845, K 124850, K 128713, K 128786, K 129058, K 131991, K 133046, K 138136, K 143460, K 143477, 2020-2.2.1-ED-2021-00181, and TKP2021-NKTA-64 (Hungary); the Council of Science and Industrial Research, India; the Latvian Council of Science; the Ministry of Education and Science, project no. 2022/WK/14, and the National Science Center, contracts Opus 2021/41/B/ST2/01369 and 2021/43/B/ST2/01552 (Poland); the Fundacao para a Ciencia e a Tecnologia, grant CEECIND/01334/2018 (Portugal); the National Priorities Research Program by Qatar National Research Fund; MCIN/AEI/10.13039/501100011033, ERDF "a way of making Europe", and the Programa Estatal de Fomento de la Investigacion Cientifica y Tecnica de Excelencia Maria de Maeztu, grant MDM-2017-0765 and Programa Severo Ochoa del Principado de Asturias (Spain); the Chulalongkorn Academic into Its 2nd Century Project Advancement Project, and the National Science, Research and Innovation Fund via the Program Management Unit for Human Resources & Institutional Development, Research and Innovation, grant B05F650021 (Thailand); the Kavli Foundation; the Nvidia Corporation; the SuperMicro Corporation; the Welch Foundation, contract C-1845; and the Weston Havens Foundation (USA).Publisher versio
Effect of different surface treatments on retention of cement-retained, implant-supported crowns
No full textTo evaluate the effect of different treatments applied to titanium implant abutment surfaces on the retention of implant-supported crowns retained using resin cement. Materials and Methods: A total of 72 titanium implant abutments were divided into six groups (n = 12 each) based on the selected surface treatment: (1) untreated; (2) sandblasted; (3) hydrogen peroxide–etched; (4) atmospheric plasma; (5) chemical mechanical polishing; and (6) titanium dioxide nanocoating. After the surface treatments, scanning electron microscopy analyses and roughness measurements of the abutment surfaces were performed. Then, 72 metal copings were fabricated and cemented onto the abutments with dual-curing resin cement. After the thermocycling process, crown retention was measured by using a universal testing machine. The experimental results were statistically evaluated with one-way analysis of variance, Tukey honest significant difference, and Tamhane T2 tests. Results: The highest surface roughness values were obtained in the sandblasted group (1.44 um), which also showed in the highest retention values (828.5 N), followed by the hydrogen peroxide etching group (490.7 N), the atmospheric plasma group (466.5 N), the chemical mechanical polishing group (410.8 N), and the control group (382.6 N). Conclusion: Sandblasting, hydrogen peroxide, etching, and atmospheric plasma treatments significantly increased crown retention, and all alternative treatments, with the exception of TiO2 nanocoating, worked better than the control samples
Performance analysis of meta-learning based bayesian deep kernel transfer methods for regression tasks
No full textMeta-learning aims to apply existing models on new tasks where the goal is 'learning to learn' so that learning from a limited amount of labeled data or learning in a short amount of time is possible. Deep Kernel Transfer (DKT) is a recently proposed meta-learning approach based on Bayesian framework. DKT's performance depends on the used kernel functions and it has two implementations, namely DKT and GPNet. In this paper, we use a large set of kernel functions on both DKT and GPNet implementations for two regression tasks to study their performances and train them under different optimizers. Furthermore, we compare the training time of both implementations to clarify the ambiguity in terms of which algorithm runs faster for the regression based tasks
Çatışmalarla açık uçlu kutulama problemi
No full textIn this thesis study, we focus on a new variant of the famous Bin Packing Problem (BPP) called the Open-End Bin Packing Problem with Conflicts (OEBPPC) which combines the Open-End Bin Packing Problem (OEBPP) and the Bin Packing Problem with Conflicts (BPPC). In OEBPPC, the aim is to pack a set of items into the least number of bins. However, the bin capacity is allowed to be exceeded only by the last item packed into the bin, and there exist conflicts between some item pairs; they cannot be packed into the same bin. We introduce a mathematical formulation and propose lower bounding procedures for our problem. We propose a metaheuristic algorithm, namely Variable Neighborhood Search (VNS), to approach the optimal solution through systematic changes and improvements in the solution. We generate different sets of instances by adapting some instances from the literature to our problem. We compare the performance of our metaheuristic algorithm both against the best lower bound and other algorithms we adapted from the literature as benchmark algorithms. We observe that our proposed metaheuristic outperforms the best benchmark algorithm in 74% of the instances with varying features.Bu tez ̧calışmasında, Açık Uçlu Kutulama Problemi (AUKP) ve Çatışmalarla Kutulama Problemi (ÇKP)'ni birleştiren, ünlü Kutulama Problemi (KP)'nin yeni bir ̧çeşidi olan Çatışmalarla Açık Uçlu Kutulama Problemi (ÇAUKP)'ne odaklanıyoruz. ÇAUKP'de amaç, bir dizi eşyayı en az sayıda kutuya paketlemektir. Ancak, kutu kapasitesinin yalnızca kutuya paketlenen son eşya tarafından aşılmasına izin verilir ve bazı eşya çiftleri arasında çelişkiler vardır; bunlar aynı kutuya paketlenemezler. Problemimiz için matematiksel bir formülasyon sunuyoruz ve alt sınır bulma yöntemleri öneriyoruz. Çözümdeki sistematik değişiklikler ve iyileştirmelerle en iyi çözüme yaklaşmak için Değişken Komşuluk Arama (DKA) adlı metasezgisel bir algoritma öneriyoruz. Literatürdeki bazı örnekleri problemimize uyarlayarak farklı örnek kümeleri oluşturuyoruz. Metasezgisel algoritmamızın performansını hem en iyi alt sınırla hem de literatürden kıyaslama algoritmaları olarak uyarladığımız diğer algoritmalarla karşılaştırıyoruz. Önerilen metasezgiselimizin, değişen özelliklere sahip örneklerin %74'ünde en iyi kıyaslama algoritmasından daha iyi performans gösterdiği gözlemlenmektedir
Teaching self-regulation through role modeling in K-12
No full textFor K-12 teachers to develop effective teaching skills, integration of role modeling strategies into teaching and learning process as a dimension of selfregulated learning is of the foremost value. Role modeling strategy training through a hybrid professional development model bears the potential to serve as a facilitating component in promoting K-12 teachers’ instructional competence. Conducted within the self-regulated learning framework, this study suggested findings of a teacher professional development training aimed at role modeling strategy implementation at K-12 level. Pursuing a mixed-method model, the current research was performed with 16 teachers who were trained and supervised to integrate role-modeling strategies into their teaching context. In this study, the data sources were role-modeling- integrated lesson plans, trainers’ feedback on these lesson plans, and online student products. The data collection methods included lesson plan evaluation through a role-modeling rubric in a quantitative fashion, whereas content analysis of trainer feedback on lesson plans, latterly revised lesson plans and online student products composed the qualitative aspect. Results revealed that this professional development training achieved significantly positive changes in teachers’ role modeling strategy implementation skills, particularly in terms of teachers’ role as agents in students’ self-regulated learning skills, promotion of student-centered learning and overall improvement in students’ self-regulated learning skills. Further, the integration of education technology tools into lessons was observed to have a positive impact on enhancing students’ self-regulated learning skills. This study could offer major contributions to designing teacher professional development training for researchers, practitioners, and teacher trainers, particularly in role modeling dimension of self-regulated learning. Copyrigh
Tasarım ve karakterizasyonu sürekli glikoz izleme sensörü
No full textThe main objective of this thesis is the development and the elaboration of a low cost continuous glucose monitor. This thesis describes the design of miniaturized continuous glucose monitoring sensor based on two needles. The sensor is based on enzymatic detection of glucose. The used enzyme is glucose oxidase entrapped inside a protecting film of chitosan. Chapter I provides a general introduction to diabetes and the different types of biosensors used for continuous glucose moni toring. “Continuous” monitoring approaches involve subcutaneous implantation of a micro-needle allowing real-time glucose level measurements in the intersti tial fluid. The working principals of continuous glucose monitors (CGM)s are introduced. An overview of electrochemical sensors is presented, and the differ ent methods of common electrode modification are also discussed in this chapter. The electrical circuits that are used for glucose sensors are reviewed. The second chapter contains the methods and materials used in the performed experiments. It describes the steps of sensor elaboration from the electrical circuit assembly on the PCB board to the electrode modification process. The measurement setup is also presented in this chapter. Chapter 3 discusses the results obtained for the sensor tests. Chapter 4 is the conclusion, it stipulates that after the tests carried on the elaborated biosensor both in the buffer solution and in the real sample, the sensor can be improved and it has potential to be used in tests on humans.Bu tezin temel amacı, d¨u¸s¨uk maliyetli bir s¨urekli glikoz monit¨or¨un¨un geli¸stirilmesi ve detaylandırılmasıdır. Bu tez, iki i˘gneye dayalı minyat¨ur s¨urekli glikoz izleme sens¨or¨un¨un tasarımını a¸cıklamaktadır. Sens¨or, glikozun enzimatik tespitine dayan maktadır. Kullanılan enzim, koruyucu bir kitosan filmi i¸cinde tutulan glikoz oksi dazdır. B¨ol¨um I, diyabete ve s¨urekli glikoz izleme i¸cin kullanılan farklı biyosens¨or t¨urlerine genel bir giri¸s sa˘glar. “S¨urekli” izleme yakla¸sımları, interstisyel sıvıdaki glikoz konsantrasyonlarının ger¸cek zamanlı ¨ol¸c¨um¨une izin veren bir mikroi˘gnenin subkutan implantasyonunu i¸cerir. S¨urekli glikoz monit¨orlerinin (CGM) ¸calı¸sma prensipleri tanıtılmaktadır. Elektrokimyasal sens¨orlere genel bir bakı¸s sunulur ve bu b¨ol¨umde yaygın elektrot modifikasyonunun farklı y¨ontemleri de tartı¸sılır. Glikoz sens¨orleri i¸cin kullanılan elektrik devreleri g¨ozden ge¸cirilir. ˙Ikinci b¨ol¨um, ger¸cekle¸stirilen deneylerde kullanılan y¨ontem ve malzemeleri i¸cermektedir. PCB kartındaki elektrik devresi montajından elektrot modifikasyon i¸slemine kadar sens¨or detaylandırma adımlarını a¸cıklar. Ol¸c¨um kurulumu da bu b¨ol¨umde sunulmaktadır. ¨ B¨ol¨um 3, sens¨or testleri i¸cin elde edilen sonu¸cları tartı¸sır. B¨ol¨um 4 sonu¸ctur, ayrıntılı biyosens¨or ¨uzerinde hem tampon ¸c¨ozeltide hem de ger¸cek numunede yapılan testlerden sonra sens¨or¨un geli¸stirilebilece˘gini ve insanlar ¨uzerinde testlerde kullanılma potansiyeline sahip oldu˘gunu ¸sart ko¸sar
Measurement of the top quark pole mass using t¯t+jet events in the dilepton final state in proton-proton collisions at √s = 13 TeV
No full textA measurement of the top quark pole mass m(t)(pole) in events where a top quarkantiquark pair (t (t) over bar) is produced in association with at least one additional jet ( t (t) over bar +jet) is presented. This analysis is performed using proton-proton collision data at root s = 13TeV collected by the CMS experiment at the CERN LHC, corresponding to a total integrated luminosity of 36.3 fb(-1). Events with two opposite-sign leptons in the final state (e(+) e(-), mu(+) mu(-), e(+/-) mu(-/+)) are analyzed. The reconstruction of the main observable and the event classification are optimized using multivariate analysis techniques based on machine learning. The production cross section is measured as a function of the inverse of the invariant mass of the t t +jet system at the parton level using a maximum likelihood unfolding. Given a reference parton distribution function (PDF), the top quark pole mass is extracted using the theoretical predictions at next-to-leading order. For the ABMP16NLO PDF, this results in m(t)(pole) = 172.93 +/- 1.36 GeV.BMBWF and FWF (Austria); FNRS and FWO (Belgium); CNPq, CAPES, FAPERJ, FAPERGS, and FAPESP (Brazil); MES and BNSF (Bulgaria); CERN; CAS, MoST, and NSFC (China); MINCIENCIAS (Colombia); MSES and CSF (Croatia); RIF (Cyprus); SENESCYT (Ecuador); MoER, ERC PUT and ERDF (Estonia); Academy of Finland, MEC, and HIP (Finland); CEA and CNRS/IN2P3 (France); BMBF, DFG, and HGF (Germany); GSRI (Greece); NKFIH (Hungary); DAE and DST (India); IPM (Iran); SFI (Ireland); INFN (Italy); MSIP and NRF (Republic of Korea); MES (Latvia); LAS (Lithuania); MOE and UM (Malaysia); BUAP, CINVESTAV, CONACYT, LNS, SEP, and UASLP-FAI (Mexico); MOS (Montenegro); MBIE (New Zealand); PAEC (Pakistan); MES and NSC (Poland); FCT (Portugal); MESTD (Serbia); MCIN/AEI and PCTI (Spain); MOSTR (Sri Lanka); Swiss Funding Agencies (Switzerland); MST (Taipei); MHESI and NSTDA (Thailand); TUBITAK and TENMAK (Turkey); NASU (Ukraine); STFC (United Kingdom); DOE and NSF (USA).r Individuals have received support from the Marie-Curie programme and the European Research Council and Horizon 2020 Grant, contract Nos. 675440, 724704, 752730, 758316, 765710, 824093, 884104, and COST Action CA16108 (European Union); the Leventis Foundation; the Alfred P. Sloan Foundation; the Alexander von Humboldt Foundation; the Belgian Federal Science Policy Office; the Fonds pour la Formation a la Recherche dans l'Industrie et dans l'Agriculture (FRIA-Belgium); the Agentschap voor Innovatie door Wetenschap en Technologie (IWT-Belgium); the F.R.S.-FNRS and FWO (Belgium) under the "Excellence of Science -EOS" -be.h project n. 30820817; the Beijing Municipal Science & Technology Commission, No. Z191100007219010; the Ministry of Education, Youth and Sports (MEYS) of the Czech Republic; the Hellenic Foundation for Research and Innovation (HFRI), Project Number 2288 (Greece); the Deutsche Forschungsgemeinschaft (DFG), under Germany's Excellence Strategy -EXC 2121 "Quantum Universe" -390833306, and under project number 400140256 -GRK2497; the Hungarian Academy of Sciences, the New National Excellence Program -UNKP, the NKFIH research grants K 124845, K 124850, K 128713, K 128786, K 129058, K 131991, K 133046, K 138136, K 143460, K 143477, 2020-2.2.1-ED-2021-00181, and TKP2021-NKTA-64 (Hungary); the Council of Science and Industrial Research, India; the Latvian Council of Science; the Ministry of Education and Science, project no. 2022/WK/14, and the National Science Center, contracts Opus 2021/41/B/ST2/01369 and 2021/43/B/ST2/01552 (Poland); the Fundacao para a Ciencia e a Tecnologia, grant CEECIND/01334/2018 (Portugal); the National Priorities Research Program by Qatar National Research Fund; MCIN/AEI/10.
13039/501100011033, ERDF "a way of making Europe", and the Programa Estatal de Fomento de la Investigacion Cientifica y Tecnica de Excelencia Maria de Maeztu, grant MDM-2017-0765 and Programa Severo Ochoa del Principado de Asturias (Spain); the Chulalongkorn Academic into Its 2nd Century Project Advancement Project, and the National Science, Research and Innovation Fund via the Program Management Unit for Human Resources & Institutional Development, Research and Innovation, grant B05F650021 (Thailand); the Kavli Foundation; the Nvidia Corporation; the SuperMicro Corporation; the Welch Foundation, contract C-1845; and the Weston Havens Foundation (USA).Publisher versio