Bibliographies thématiques / Tiny ML

Sommaire

Articles de revues
Thèses
Chapitres de livres
Actes de conférences

Littérature scientifique sur le sujet « Tiny ML »

Auteur : Grafiati

Publié le 9 mars 2023

Mis à jour le 10 mars 2023

Créez une référence correcte selon les styles APA, MLA, Chicago, Harvard et plusieurs autres

Choisissez une source :

Consultez les listes thématiques d’articles de revues, de livres, de thèses, de rapports de conférences et d’autres sources académiques sur le sujet « Tiny ML ».

À côté de chaque source dans la liste de références il y a un bouton « Ajouter à la bibliographie ». Cliquez sur ce bouton, et nous générerons automatiquement la référence bibliographique pour la source choisie selon votre style de citation préféré : APA, MLA, Harvard, Vancouver, Chicago, etc.

Vous pouvez aussi télécharger le texte intégral de la publication scolaire au format pdf et consulter son résumé en ligne lorsque ces informations sont inclues dans les métadonnées.

Articles de revues sur le sujet "Tiny ML"

Lahade, Shashikant Vitthalrao, Srikanth Namuduri, Himanshu Upadhyay et Shekhar Bhansali. « Alcohol Sensor Calibration on the Edge Using Tiny Machine Learning (Tiny-ML) Hardware ». ECS Meeting Abstracts MA2020-01, n^o 26 (1 mai 2020) : 1848. http://dx.doi.org/10.1149/ma2020-01261848mtgabs.

Texte intégral

Styles APA, Harvard, Vancouver, ISO, etc.

Lulec, Sevil Zeynep, Alvin Loke, Xinfei Guo, Ka-Meng Lei, Po-Hsuan Wei, Shahriar Mirabbasi, Abira Altvater et Kelsey Rodriguez. « IEEE SSCS and tiny ML Hold First Young Professionals Webinar [Chapters] ». IEEE Solid-State Circuits Magazine 12, n^o 3 (2020) : 51–53. http://dx.doi.org/10.1109/mssc.2020.3001885.

Texte intégral

Styles APA, Harvard, Vancouver, ISO, etc.

Jose Rosa, Daniel Granhao, Guilherme Carvalho, Tiago Gon�alves, Monica Figueiredo, Luis Conde Bento, Nuno Paulino et Luis M. Pessoa. « BacalhauNet : A tiny CNN for lightning-fast modulation classification ». ITU Journal on Future and Evolving Technologies 3, n^o 2 (22 septembre 2022) : 252–60. http://dx.doi.org/10.52953/fywt4006.

Texte intégral

Résumé :

Deep learning methods have been shown to be competitive solutions for modulation classification tasks, but suffer from being computationally expensive, limiting their use on embedded devices. We propose a new deep neural network architecture which employs known structures, depth-wise separable convolution and residual connections, as well as a compression methodology, which combined lead to a tiny and fast algorithm for modulation classification. Our compressed model won the first place in ITU's AI/ML in 5G Challenge 2021, achieving 61.73� compression over the challenge baseline and being over 2.6� better than the second best submission. The source code of this work is publicly available at github.com/ITU-AI- ML-in-5G-Challenge/ITU-ML5G-PS-007-BacalhauNet.

Styles APA, Harvard, Vancouver, ISO, etc.

ZAROMB, SOLOMON, DENNIS MARTELL, NATHAN SCHATTKE et GARY HANKINS. « PRECONCENTRATION OF MICROORGANISMS INTO A TINY VOLUME OF LIQUID FOR ENHANCED SPECTRAL DETECTION ». International Journal of High Speed Electronics and Systems 17, n^o 04 (décembre 2007) : 739–46. http://dx.doi.org/10.1142/s0129156407004941.

Texte intégral

Résumé :

Concentration of 1-micron-size micro-organisms from about 100 ml of liquid, whether drawn from a bio-aerosol collector or from an environmental water source, into a volume of 1 to 2 ml. is achieved by a liquid flow system including a reversible filter through which filtered liquid can be recirculated or disposed of and from which a concentrated sample is recovered by opening a solenoid valve leading to a detector or to a collection container and reversing the pump for a short time. The reversing action flushes the collected particles off the filter and into the detector or container. The effectiveness of this approach was demonstrated by measuring the cumulative fluorescence of 1-micron-size blue fluorescent microspheres versus cumulative volume withdrawn from our WEP collector after capture from an aerosol suspension of about 140 particles/ml drawn from a test chamber over a 5-minute period at a sampling rate of 500 liters/minute. As the liquid was being withdrawn from the WEP collector, it was filtered at a rate of 1 ml/second and the filter back-flushed with small portions of filtrate at 1-minute intervals. The relative cumulative concentration of captured particles in the first back-flushed 1-ml fraction was around 56 as compared with a value of 1.4 in the first 60-ml filtered fraction, which constitutes a liquid-to-liquid concentration enhancement by a factor of 40 and an air-to-liquid concentration factor of >1.25 × 106.

Styles APA, Harvard, Vancouver, ISO, etc.

Sozzi, Marco, Giulio Pillan, Claudia Ciarelli, Francesco Marinello, Fabrizio Pirrone, Francesco Bordignon, Alessandro Bordignon, Gerolamo Xiccato et Angela Trocino. « Measuring Comfort Behaviours in Laying Hens Using Deep-Learning Tools ». Animals 13, n^o 1 (21 décembre 2022) : 33. http://dx.doi.org/10.3390/ani13010033.

Texte intégral

Résumé :

Image analysis using machine learning (ML) algorithms could provide a measure of animal welfare by measuring comfort behaviours and undesired behaviours. Using a PLF technique based on images, the present study aimed to test a machine learning tool for measuring the number of hens on the ground and identifying the number of dust-bathing hens in an experimental aviary. In addition, two YOLO (You Only Look Once) models were compared. YOLOv4-tiny needed about 4.26 h to train for 6000 epochs, compared to about 23.2 h for the full models of YOLOv4. In validation, the performance of the two models in terms of precision, recall, harmonic mean of precision and recall, and mean average precision (mAP) did not differ, while the value of frame per second was lower in YOLOv4 compared to the tiny version (31.35 vs. 208.5). The mAP stands at about 94% for the classification of hens on the floor, while the classification of dust-bathing hens was poor (28.2% in the YOLOv4-tiny compared to 31.6% in YOLOv4). In conclusion, ML successfully identified laying hens on the floor, whereas other PLF tools must be tested for the classification of dust-bathing hens.

Styles APA, Harvard, Vancouver, ISO, etc.

Sudharsan, Bharath. « Training Up to 50 Class ML Models on 3 $ IoT Hardware via Optimizing One-vs-One Algorithm (Student Abstract) ». Proceedings of the AAAI Conference on Artificial Intelligence 36, n^o 11 (28 juin 2022) : 13059–60. http://dx.doi.org/10.1609/aaai.v36i11.21666.

Texte intégral

Résumé :

Multi-class classifier training using traditional meta-algorithms such as the popular One-vs-One (OvO) method may not always work well under cost-sensitive setups. Also, during inference, OvO becomes computationally challenging for higher class counts K as O(K^2) is its time complexity. In this paper, we present Opt-OvO, an optimized (resource-friendly) version of the One-vs-One algorithm to enable high-performance multi-class ML classifier training and inference directly on microcontroller units (MCUs). Opt-OvO enables billions of tiny IoT devices to self learn/train (offline) after their deployment, using live data from a wide range of IoT use-cases. We demonstrate Opt-OvO by performing live ML model training on 4 popular MCU boards using datasets of varying class counts, sizes, and feature dimensions. The most exciting finding was, on the 3 $ ESP32 chip, Opt-OvO trained a multi-class ML classifier using a dataset of class count 50 and performed unit inference in super real-time of 6.2 ms.

Styles APA, Harvard, Vancouver, ISO, etc.

Farag, Mohammed M. « A Tiny Matched Filter-Based CNN for Inter-Patient ECG Classification and Arrhythmia Detection at the Edge ». Sensors 23, n^o 3 (26 janvier 2023) : 1365. http://dx.doi.org/10.3390/s23031365.

Texte intégral

Résumé :

Automated electrocardiogram (ECG) classification using machine learning (ML) is extensively utilized for arrhythmia detection. Contemporary ML algorithms are typically deployed on the cloud, which may not always meet the availability and privacy requirements of ECG monitoring. Edge inference is an emerging alternative that overcomes the concerns of cloud inference; however, it poses new challenges due to the demanding computational requirements of modern ML algorithms and the tight constraints of edge devices. In this work, we propose a tiny convolutional neural network (CNN) classifier for real-time monitoring of ECG at the edge with the aid of the matched filter (MF) theory. The MIT-BIH dataset with inter-patient division is used for model training and testing. The model generalization capability is validated on the INCART, QT, and PTB diagnostic databases, and the model performance in the presence of noise is experimentally analyzed. The proposed classifier can achieve average accuracy, sensitivity, and F1 scores of 98.18%, 91.90%, and 92.17%, respectively. The sensitivity of detecting supraventricular and ventricular ectopic beats (SVEB and VEB) is 85.3% and 96.34% , respectively. The model is 15KB in size, with an average inference time of less than 1 ms. The proposed model achieves superior classification and real-time performance results compared to the state-of-the-art ECG classifiers while minimizing the model complexity. The proposed classifier can be readily deployed on a wide range of resource-constrained edge devices for arrhythmia monitoring, which can save millions of cardiovascular disease patients

Styles APA, Harvard, Vancouver, ISO, etc.

Toyama, Haruko, Hong Rui Huang, Tomonori Nakamura, Leonid V. Bondarenko, Alexandra Y. Tupchaya, Dimitry V. Gruznev, Akari Takayama, Andrey V. Zotov, Aleksandr A. Saranin et Shuji Hasegawa. « Superconductivity of Pb Ultrathin Film on Ge(111) Surface ». Defect and Diffusion Forum 386 (septembre 2018) : 80–85. http://dx.doi.org/10.4028/www.scientific.net/ddf.386.80.

Texte intégral

Résumé :

We have performed structure analysis and electrical conductivity measurements of Pb ultrathin films of different thicknesses grown on Ge (111) at low temperature by using electron diffraction, scanning tunneling microscopy, andin-situfour-point probe method in ultrahigh vacuum. Three samples with different deposition amounts of Pb corresponding to 1, 3 and 10 monolayer (ML) were revealed to have different structures. The 1 ML-Pb sample, having a wetting layer and tiny clusters on it, did not show superconductivity. The 10-ML-Pb sample, consisted of continuous Pb (111) thin film structure, showed thin-film superconductivity around 6 K. The 3-ML-Pb sample, consisted of the wetting layer with unconnected Pb (111) islands on it, also showed superconductivity around 4 K. This superconductivity is thought to be induced in the wetting layer by proximity effect from superconducting Pb (111) islands. Thus, it is important to study the detailed growth structures for understanding atomic-layer superconductivity.

Styles APA, Harvard, Vancouver, ISO, etc.

Mao, Jian, Long Wang, Zhiyong Qian et Mingjing Tu. « Uptake and Cytotoxicity of Ce(IV) Doped TiO2Nanoparticles in Human Hepatocyte Cell Line L02 ». Journal of Nanomaterials 2010 (2010) : 1–8. http://dx.doi.org/10.1155/2010/910434.

Texte intégral

Résumé :

Ce(IV) doped anatase TiO2nanoparticles (CDTs) were prepared and the underlying mechanism by which CDT nanoparticle enters into cell and its cytotoxicity were investigated in a human hepatocellular line L02 cell. The results showed that CDTs can enter into cytoplasm of L02 cell via endocytosis and nonendocytic ways. Large aggregation of CDTs went into cell by endocytosis and finally formed an endocytic vesicle with membrane boundary. Tiny aggregation of CDTs entered into cell cytoplasm via channels similar to that for lung-blood substance exchange in the alveolar-airway barrier. In addition, tiny aggregation of CDTs was observed in nucleus, and maybe CDTs could pass through the nucleus envelope via the channels provided by nuclear pore complexes (NPCs). Results from MTT assay, fluorescence microscope, and TEM observations showed that the cell viability, cell morphology, cell growth, and cell division periods could not be obviously impaired when cells were exposed to CDTs of different concentration from 30 to 150 μg mL−1without UV irradiation. However, large vacuoles containing CDTs were found in cytoplasm, some structure changes were observed in mitochondria, and smooth envelope around the nucleus was shrank and deformed.

Styles APA, Harvard, Vancouver, ISO, etc.

Golla, Kishore, et S. PallamSetty. « An Efficient Secure Cryptography Scheme for New ML-based RPL Routing Protocol in Mobile IoT Environment ». International Journal of Network Security & ; Its Applications 14, n^o 2 (31 mars 2022) : 1–13. http://dx.doi.org/10.5121/ijnsa.2022.14201.

Texte intégral

Résumé :

Internet of Things (IoT) offers reliable and seamless communication for the heterogeneous dynamic lowpower and lossy network (LLNs). To perform effective routing in IoT communication, LLN Routing Protocol (RPL) is developed for the tiny nodes to establish connection by using deflaut objective functions: OF0, MRHOF, for which resources are constraints like battery power, computation capacity, memory communication link impacts on varying traffic scenarios in terms of QoS metrics like packet delivery ratio, delay, secure communication channel. At present, conventional Internet of Things (IoT) are having secure communication channels issue for transmission of data between nodes. To withstand those issues, it is necessary to balance resource constraints of nodes in the network. In this paper, we developed a security algorithm for IoT networks with RPL routing. Initially, the constructed network in corporates optimizationbased deep learning (reinforcement learning) for route establishment in IoT. Upon the establishment of the route, the ClonQlearn based security algorithm is implemented for improving security which is based onaECC scheme for encryption and decryption of data. The proposed security technique incorporates reinforcement learning-based ClonQlearnintegrated with ECC (ClonQlearn+ECC) for random key generation. The proposed ClonQlearn+ECCexhibits secure data transmission with improved network performance when compared with the earlier works in simulation. The performance of network expressed that the proposed ClonQlearn+ECC increased the PDR of approximately 8% - 10%, throughput of 7% - 13%, end-to-end delay of 5% - 10% and power consumption variation of 3% - 7%.

Styles APA, Harvard, Vancouver, ISO, etc.

Plus de sources

Thèses sur le sujet "Tiny ML"

PEREGO, RICCARDO. « Automated Deep Learning through Constrained Bayesian Optimization ». Doctoral thesis, Università degli Studi di Milano-Bicocca, 2021. http://hdl.handle.net/10281/314922.

Texte intégral

Résumé :

In un mondo sempre più tecnologico e interconnesso, la quantità di dati è in continua crescita e, di conseguenza, anche gli algoritmi di decision-making sono in continua evoluzione per adattarsi ad essi. Una delle principali fonti di questa grande quantità di dati è l'Internet of Things, in cui miliardi di sensori si scambiano informazioni attraverso la rete per svolgere vari tipi di attività come il monitoraggio industriale e medico. Negli ultimi anni lo sviluppo tecnologico ha permesso di definire nuove architetture hardware ad alte prestazioni per i sensori, detti microcontrollori, che hanno permesso la creazione di un nuovo tipo di calcolo decentralizzato denominato Edge Computing. Questo nuovo paradigma di calcolo ha permesso ai sensori di eseguire algoritmi di decision-making per prendere decisioni immediate e locali invece di trasferire i dati su un server centrale di elaborazione. Per supportare l'Edge Computing, la comunità di ricerca ha iniziato a sviluppare nuove tecniche avanzate per gestire in modo efficiente le limitate risorse su questi dispositivi per l'applicazione dei più avanzati modelli di Machine Learning, in particolare le Deep Neural Network. Automated Machine Learning è una branca del campo del Machine Learning che mira a divulgare la potenza del Machine Learning ai non esperti, oltre a supportare in modo efficiente i data scientists nella progettazione delle proprie pipeline di analisi dei dati. L'adozione del Automated Machine Learning ha reso possibile lo sviluppo quasi automatico di modelli sempre più performanti. Tuttavia, con l'avvento dell'Edge Computing, è nata una specializzazione del Machine Learning, definita come Tiny Machine Learning (Tiny ML), ovvero l'applicazione di algoritmi di Machine Learning su dispositivi con risorse hardware limitate. Questa tesi si occupa principalmente dell'applicabilità del Automated Machine Learning per generare modelli accurati che devono essere anche implementabili su dispositivi minuscoli, in particolare i microcontrollori. Più specificamente, l'approccio proposto è volto a massimizzare le prestazioni delle Reti Neurali Profonde e a soddisfare i vincoli associati alle limitate risorse hardware, comprese le batterie, dei microcontrollori. Grazie ad una stretta collaborazione con STMicroelectronics, azienda leader nella progettazione, produzione e vendita di microcontrollori, è stato possibile sviluppare un nuovo framework di Automated Machine Learning che si occupa dei vincoli black-box relativi all’impossibilità di implementare una Deep Neural Network su questi piccoli dispositivi, ampiamente adottato nelle applicazioni dell'internet degli oggetti. L'applicazione su due casi d'uso reali forniti da STMicroelectronics (ad esempio, Human Activity Recognition e Image Recognition) ha dimostrato che il nuovo approccio proposto è in grado di trovare in modo efficiente configurazioni per reti neurali profonde accurate e implementabili, aumentandone l'accuratezza rispetto ai modelli di base e riducendo drasticamente le risorse hardware necessarie per farle funzionare su un microcontrollore (cioè, una riduzione di oltre il 90%). L'approccio è stato anche confrontato con una delle soluzioni AutoML all'avanguardia per valutare la sua capacità di superare i problemi che attualmente limitano l'ampia applicazione di AutoML nel campo del Tiny ML. Infine, questa tesi di dottorato suggerisce interessanti e stimolanti direzioni di ricerca per aumentare ulteriormente l'applicabilità dell'approccio proposto, integrando i risultati di ricerche recenti e innovative (ad esempio, weakly defined search spaces, Meta-Learning, Multi-objective and Multi-Information Source optimization).
In an increasingly technological and interconnected world, the amount of data is continuously growing, and as a consequence, decision-making algorithms are also continually evolving to adapt to it. One of the major sources of this vast amount of data is the Internet of Things, in which billions of sensors exchange information over the network to perform various types of activities such as industrial and medical monitoring. In recent years, technological development has made it possible to define new high-performance hardware architectures for sensors, called Microcontrollers, which enabled the creation of a new kind of decentralized computing named Edge Computing. This new computing paradigm allowed sensors to run decision-making algorithm at the edge in order to take immediate and local decisions instead of transferring the data on central server processing. To support Edge Computing, the research community started developing new advanced techniques to efficiently manage the limited resources on these devices for applying the most advanced Machine Learning models, especially the Deep Neural Networks. Automated Machine Learning is a branch of the Machine Learning field aimed at disclosing the power of Machine Learning to non-experts as well as efficiently supporting data scientists in designing their own data analysis pipelines. The adoption of Automated Machine Learning has made it possible to develop increasingly high-performance models almost automatically. However, with the advent of Edge Computing, a specialization of Machine Learning, defined as Tiny Machine Learning (Tiny ML), has been arising, that is, the application of Machine Learning algorithms on devices having limited hardware resources. This thesis mainly addresses the applicability of Automated Machine Learning to generate accurate models which must be also deployable on tiny devices, specifically Microcontroller Units. More specifically, the proposed approach is aimed at maximizing the performances of Deep Neural Networks while satisfying the constraints associated to the limited hardware resources, including batteries, of Microcontrollers. Thanks to a close collaboration with STMicroelectronics, a leading company for design, production and sale of microcontrollers, it was possible to develop a novel Automated Machine Learning framework that deals with the black-box constraints related to the deployability of a Deep Neural Network on these tiny devices, widely adopted in IoT applications. The application on two real-life use cases provided by STMicroelectronics (i.e., Human Activity Recognition and Image Recognition) proved that the novel proposed approach can efficiently find out configurations for accurate and deployable Deep Neural Networks, increasing their accuracy against baseline models while drastically reducing hardware required to run them on a microcontroller (i.e., a reduction of more than 90\%). The approach was also compared against one of the state-of-the-art AutoML solutions in order to evaluate its capability to overcome the issues which currently limit the wide application of AutoML in the tiny ML field. Finally, this PhD thesis suggests interesting and challenging research directions to further increase the applicability of the proposed approach by integrating recent and innovative research results (e.g., weakly defined search spaces, Meta-Learning, Multi-objective and Multi-Information Source optimization).

Styles APA, Harvard, Vancouver, ISO, etc.

Chapitres de livres sur le sujet "Tiny ML"

Gutti, Vivek, et R. Karthi. « Real Time Classification of Fruits and Vegetables Deployed on Low Power Embedded Devices Using Tiny ML ». Dans Third International Conference on Image Processing and Capsule Networks, 347–59. Cham : Springer International Publishing, 2022. http://dx.doi.org/10.1007/978-3-031-12413-6_27.

Texte intégral

Styles APA, Harvard, Vancouver, ISO, etc.

Funk, Frederik, Thorsten Bucksch et Daniel Mueller-Gritschneder. « ML Training on a Tiny Microcontroller for a Self-adaptive Neural Network-Based DC Motor Speed Controller ». Dans Communications in Computer and Information Science, 268–79. Cham : Springer International Publishing, 2020. http://dx.doi.org/10.1007/978-3-030-66770-2_20.

Texte intégral

Styles APA, Harvard, Vancouver, ISO, etc.

Vuppalapati, Chandrasekar, Anitha Ilapakurti, Sharat Kedari, Raja Vuppalapati, Jaya Vuppalapati et Santosh Kedari. « Crossing the Artificial Intelligence (AI) Chasm, Albeit Using Constrained IoT Edges and Tiny ML, for Creating a Sustainable Food Future ». Dans Proceedings of Fifth International Congress on Information and Communication Technology, 540–53. Singapore : Springer Singapore, 2020. http://dx.doi.org/10.1007/978-981-15-5859-7_54.

Texte intégral

Styles APA, Harvard, Vancouver, ISO, etc.

Sharma, Avinash Kumar, Pranav Kumar Tripathi et Sushant Sharma. « Role of Artificial Intelligence in Biomedical Imaging ». Dans Advances in Medical Technologies and Clinical Practice, 17–34. IGI Global, 2023. http://dx.doi.org/10.4018/978-1-6684-6957-6.ch002.

Texte intégral

Résumé :

A trendy technique based on computer science called artificial intelligence (AI) creates software and algorithms to make machines smart and effective at carrying out activities that often call for expert human intellect. Machine learning (ML), deep learning (DL), traditional neural networks, fuzzy logic, and speech recognition are only a few of the subsets of AI that have distinctive skills and functions that might enhance the performance of contemporary medical sciences. Biomedical imaging might undergo a revolution thanks to AI, which could increase the efficiency and precision of picture processing and interpretation. Radiologists could miss tiny abnormalities that can be detected by AI systems that have been taught to spot patterns in those pictures that are challenging for humans to interpret. AI may also be used to generate customized medicine by evaluating a patient's medical pictures and other data to customize treatment regimens, as well as to enhance image processing and visualization.

Styles APA, Harvard, Vancouver, ISO, etc.

Li, Jie Jack, Chris Limberakis et Derek A. Pflum. « Reductions ». Dans Modern Organic Synthesis in the Laboratory. Oxford University Press, 2008. http://dx.doi.org/10.1093/oso/9780195187984.003.0010.

Texte intégral

Résumé :

The Barton deoxygenation (or Barton–McCombie deoxygenation) is a two-step reaction sequence for the reduction of an alcohol to an alkane. The alcohol is first converted to a methyl xanthate or thioimidazoyl carbamate. Then, the xanthate or thioimidazoyl carbamate is reduced with a tin hydride reagent under radical conditions to afford the alkane. Trialkylsilanes have also been used as the hydride source. Reviews: (a) McCombie, S. W. In Comprehensive Organic Synthesis; Trost, B. M.; Fleming, I., Eds.; Pergamon Press: Oxford, U. K., 1991; Vol. 8, Chapter 4.2: Reduction of Saturated Alcohols and Amines to Alkanes, pp. 818–824. (b) Crich, D.; Quintero, L. Chem. Rev. 1989, 89, 1413–1432. To a solution of the â-hydroxy-N-methyl-O-methylamide (0.272 g, 1.55 mol) in tetrahydrofuran (THF) (30 mL) were added carbon disulfide (6.75 mL, 112 mmol) and iodomethane (6.70 mL, 108 mmol) at 0 °C. The mixture was stirred at this temperature for 0.25 h, and then sodium hydride (60% suspension in mineral, 136.3 mg, 3.4 mmol) was added. After 20 min at 0 °C, the reaction was quenched by slow addition to 60 g of crushed ice. (Caution: hydrogen gas evolution!). The mixture was raised to room temperature and separated, and the aqueous layer was extracted with CH2Cl2 (4 × 15 mL). The combined organic extracts were dried (Na2SO4</aub>), concentrated in vacuo, and purified (SiO2, 5% EtOAc in hexanes) to afford 0.354 g (86%) of the xanthate. To a solution of the xanthate (2.95 g, 11.1 mmol) in toluene (100 mL) was added tributyltin hydride (15.2 mL, 56.6 mmol) and 2,2´-azobisisobutyronitrile (AIBN, 0.109 g, 0.664 mmol). The reaction mixture was then heated to reflux for 1 h. The mixture was cooled, concentrated in vacuo, and purified (SiO2, 100% hexanes to remove tin byproducts, followed by 10% EtOAc in hexanes to elute product) to afford 1.69 g (96%) of the N-methyl-O-methylamide.

Styles APA, Harvard, Vancouver, ISO, etc.

Manikkampatti Palanisamy, Murugesan, Akilamudhan Palaniappan, VenkataRatnam Myneni, Padmapriya Veerappan et Minar Mohamed Lebba. « Leaching Technology for Precious Heavy Metal Recapture through (HCI + HNO₃) and (HCI + H₂SO₄) from E-Waste ». Dans Heavy Metals - New Insights [Working Title]. IntechOpen, 2022. http://dx.doi.org/10.5772/intechopen.102347.

Texte intégral

Résumé :

The rapid growth of information technology and industrialization are the key components for the development of electronic equipment, and their inevitable role in human day-to-day life has an important stint in the generation of electronic waste (e-waste). This waste has far-reaching environmental and health consequences. One such e-waste printed circuit board (PCB) contains significant amounts of valuable heavy metals such as copper (Cu), lead (Pb), zinc (Zn), nickel (Ni), and others that can be extracted through various metallurgical routes. Recovery and recycle of heavy metal ions is a major challenge to prevent environmental contamination. The present study discusses the current e-waste scenario, health impacts and treatment methods in detail, and also presents experimental results of recovery of heavy metals from printed circuit boards (PCBs) by leaching using aqua regia (HCI + HNO3 and HCI + H2SO4). Under varying conditions such as specified conditions of 80°C, 0.05 mm of thickness, 3 hrs of contacttime, 80rpm shaking speed, and concentration of PCB sample of 0.5 g ml−1, it results in the composition of extracted heavy metal ions in such a way that 97.59% of copper, 96.59% of lead, 94.66% of tin, and 96.64% of zinc, respectively. The recovery of heavy metal ions from PCBs has an important leading contribution in electronic waste management and the result shows a higher rate.

Styles APA, Harvard, Vancouver, ISO, etc.

Actes de conférences sur le sujet "Tiny ML"

Shankar, R., K. M. Gautham Mythireyan, Nikitha Reddy Nalla et M. Venkateshkumar. « Cough Recognition Using Tiny ML ». Dans 2022 IEEE Industrial Electronics and Applications Conference (IEACon). IEEE, 2022. http://dx.doi.org/10.1109/ieacon55029.2022.9951763.

Texte intégral

Styles APA, Harvard, Vancouver, ISO, etc.

Vuppalapati, Chandrasekar, Anitha Ilapakurti, Karthik Chillara, Sharat Kedari et Vanaja Mamidi. « Automating Tiny ML Intelligent Sensors DevOPS Using Microsoft Azure ». Dans 2020 IEEE International Conference on Big Data (Big Data). IEEE, 2020. http://dx.doi.org/10.1109/bigdata50022.2020.9377755.

Texte intégral

Styles APA, Harvard, Vancouver, ISO, etc.

Nyamukuru, Maria T., et Kofi M. Odame. « Tiny Eats : Eating Detection on a Microcontroller ». Dans 2020 IEEE Second Workshop on Machine Learning on Edge in Sensor Systems (SenSys-ML). IEEE, 2020. http://dx.doi.org/10.1109/sensysml50931.2020.00011.

Texte intégral

Styles APA, Harvard, Vancouver, ISO, etc.

Probierz, Eryka, Natalia Bartosiak, Martyna Wojnar, Kamil Skowronski, Adam Galuszka, Tomasz Grzejszczak et Olaf Kedziora. « Application of Tiny-ML methods for face recognition in social robotics using OhBot robots ». Dans 2022 26th International Conference on Methods and Models in Automation and Robotics (MMAR). IEEE, 2022. http://dx.doi.org/10.1109/mmar55195.2022.9874278.

Texte intégral

Styles APA, Harvard, Vancouver, ISO, etc.

Antonini, Mattia, Miguel Pincheira, Massimo Vecchio et Fabio Antonelli. « Tiny-MLOps : a framework for orchestrating ML applications at the far edge of IoT systems ». Dans 2022 IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS). IEEE, 2022. http://dx.doi.org/10.1109/eais51927.2022.9787703.

Texte intégral

Styles APA, Harvard, Vancouver, ISO, etc.

Vuppalapati, Chandrasekar, Anitha Ilapakurti, Sharat Kedari, Jaya Vuppalapati, Santosh Kedari et Raja Vuppalapati. « Democratization of AI, Albeit Constrained IoT Devices & ; Tiny ML, for Creating a Sustainable Food Future ». Dans 2020 3rd International Conference on Information and Computer Technologies (ICICT). IEEE, 2020. http://dx.doi.org/10.1109/icict50521.2020.00089.

Texte intégral

Styles APA, Harvard, Vancouver, ISO, etc.

Ramos Gurjao, Kildare George, Eduardo Gildin, Richard Gibson et Mark Everett. « Estimation of Far-Field Fiber Optics Distributed Acoustic Sensing DAS Response Using Spatio-Temporal Machine Learning Schemes and Improvement of Hydraulic Fracture Geometric Characterization ». Dans SPE Hydraulic Fracturing Technology Conference and Exhibition. SPE, 2022. http://dx.doi.org/10.2118/209119-ms.

Texte intégral

Résumé :

Abstract Distributed Acoustic Sensing (DAS) is a fiber optics method that is revolutionizing the unconventional reservoir monitoring technology with substantial spatial coverage, high frequency data acquisition, and broad cable deployment options including hazardous/harsh environments compared to traditional geophysical methods such as point sensors (i.e., geophones). However, a single well equipped with fiber cannot acquire the far-field strain response since the sensitivity of this technique is restricted to a region near the monitor well. In this paper, we develop an Artificial Intelligence (AI) algorithm to estimate the magnitude of the far-field DAS response for any spatio-temporal input. Moreover, we identify a discontinuity in displacement results following fracture hit, which is interpreted as an effect of rock plastic deformation, and for the first time we demonstrate that it may be related to fracture width. Therefore, the output of our algorithm is used to estimate such geometric property along time in multiple locations. We generate the tangent displacement component (uy) (parallel to monitor well) using an in-house code based on Displacement Discontinuity Method (DDM). Several monitor wells are incorporated in the simulation of physical scenarios characterized by single and multiple hydraulic fractures. For each specific scenario we train and test an Artificial Neural Network (ANN) with position and time as input variables, and axial displacement as output. The Machine Learning (ML) model is designed with 7 hidden layers, 100 the maximum number of neurons per layer and hyperbolic tangent as activation function. Finally, predicted uy is used to: (1) obtain Distributed Acoustic Sensing (DAS) data deriving it sequentially in space and time; and (2) estimate fracture width based on discontinuity magnitude. Training stage is performed avoiding overfitting and minimizing ANN loss function. In the testing phase, error between true and predicted variables is negligible in the entire waterfall plot region, except at initial time steps where fracture treatment starts at operation well and magnitude of axial displacement collected at monitor well is very small on the order of 10-6 or even lower. In this case, we suspect that these tiny supervisor values may have minimal impact on the loss function, and consequently weights and biases of regression model are barely updated to consider the effect of such outputs. Regarding fracture width estimation, error reduces consistently along time at all locations reaching values near 0%. To the best of our knowledge this is the first work that creates a ML algorithm able to estimate strain fields generated during hydraulic fracturing treatments merely based on position and time inputs. The model developed with synthetic data is an incentive for the deployment of multiple monitor wells in the field to enhance beyond the near wellbore region geometric characterization of created fracture systems, and possibly identify critical patterns associated with fracture propagation that ultimately can lead to production optimization.

Styles APA, Harvard, Vancouver, ISO, etc.

Nandakumar, Krishnan, Allen G. Parr, Geon Hahm, Michael A. Huff et Stephen M. Phillips. « A Smart Shape Memory Alloy Actuated Microvalve With Feedback Control ». Dans ASME 1998 International Mechanical Engineering Congress and Exposition. American Society of Mechanical Engineers, 1998. http://dx.doi.org/10.1115/imece1998-1234.

Texte intégral

Résumé :

Abstract The microfluidics group at CWRU has demonstrated a thin film shape memory alloy actuated smart microvalve with feedback control for fluid control. A micromachined flow sensor was designed and fabricated and feedback control electronics were developed and incorporated with a TiNi microvalve to realize the system. The valve is a Titanium Nickel (TiNi) Shape Memory Alloy (SMA) thin film actuated micromachined valve capable of modulating 0–250 ml/min of airflow at 2 psi. The flow sensor used is a micromachined, anemometric type flow sensor. By simulation, the smart controller was designed using MATLAB with Simulink and realized with standard integrated circuits. The performance of the smart valve shows flow insensitivity to pressure variations across the valve. This was demonstrated by a constant flow even when the valve is subject to a change in the pressure drop across the valve. This is in contrast to the typical behavior of standard valves without flow sensing and feedback control.

Styles APA, Harvard, Vancouver, ISO, etc.

Landolfi, R., R. De Cristofaro, S. De Carolis, G. Ciabattoni et B. Bizzi. « PLACENTAL-DERIVED PGI2 INHIBITS CORD PLATELET FUNCTION : POSSIBLE ROLE OF PGI2 IN THE TRANSIENT HYPOREACTIVITY OF NEWBORN PLATELETS ». Dans XIth International Congress on Thrombosis and Haemostasis. Schattauer GmbH, 1987. http://dx.doi.org/10.1055/s-0038-1644274.

Texte intégral

Résumé :

Previous studies on newborn platelets hyporeactivity have been performed on cord blood. In this study we demonstrated that fresh cord platelet poor plasma (C-PPP) contains a labile antiaggrega-ting substance which, added to adult platelet rich plasma (PRP), is able to reverse ADP-induced platelet aggregation. Measurements of 6-Keto-prostaglandin (PG) Fl± levels in C-PPP obtained from10 different normal newborns gave anaverage value of 1050 ± 361(SD) pg/mL. Significantly lower levelsof this prostaglandin were found in plasma samples obtained from two newborns 2 hours after the birth (mean = 150 pg/mL) and in PPP of ten control adults (mean = 25 ± 34 pg/mL). In three newborns, platelet aggregation was studied using both C-PRP and PRP obtained 2 and 48 hours after the birth. A marked reductionof platelet response to ADP and collagen was evident in C-PRP. Such hyporeactivity was mild at 2 hours and absent in the third day of life. These results show that PGI2 inhibits cord platelets and might be the cause of a transient platelet hyporeactivity in the newborn. Finally we demonstrated that washed newborn platelets, compared to adult platelets, have a significant increase ofthe apparent affinity constant (Ka)for fibrinogen and that fetal and adult fibrinogen have similar Ka for platelets.

Styles APA, Harvard, Vancouver, ISO, etc.

Vila, V., E. Reganon, J. Aznar, V. Lacueva et M. Ruano. « EFFECT OF TREATMENT WITH STREPTOKINASE AND HEPARIN ON FIBRINOGEN, FIBRIN AND RELATED PROTEINS IN ACUIE MYOCARDIAL INFARCTION (/ME) PATIENTS ». Dans XIth International Congress on Thrombosis and Haemostasis. Schattauer GmbH, 1987. http://dx.doi.org/10.1055/s-0038-1644895.

Texte intégral

Résumé :

The properties of fibrinogen and fibrin, the levels of fibrincpeptide A (FPA) and fibrin(ogen) degradation products (FDP) were studied in 34 patients with AMI who were undergoing thrombolytic and heparin therapy. They were classified into 6 groups accordingto their stage of treatment: group 1, before intravenous administration of 800.000 U streptokinase over 30 min; group 2, after a<Mnistraticn of SK but before adninistraticn of heparin; group 3, during 24 h ofthe 5 ng/h heparin continuous infusion; group 4, during 48-72 h of the 16.6 ng/h heparin continuous infhsion; group 5, after 1 week of administration of SK and with a bolus inyection of 50 rg heparin every 4 h; group 6, patients who were undergoing only heparin treatment. The Fg 1/ Fg II ratio varies during treatment with SK and heparin. In group 1 a sligjnt increase (2.5) is observed. Group 2 shows a significantdecrease (0.6) as a result of fibrinolysis. In group3 the ratio reaches normal value (1.8) while in the fourth group it is twice the normal value (4). The value for group 5 is nearly normal (2.1), and in group 6 it reaches values similar to those obtained in group 4, which implies that the rise in the Fgl/Fgll ratio is not a result of fibrinolytic treatment. TheFPA level shows and increase in patients with AMI (group 1,126 ng/ml). When SK treatment is applied (group 2), FPA decreases to 52 ng/ml. Later treatment with heparin (group-3, 82; group-4, 44 and group-5, 81ng/ml) does not neutralize thrcmbinic activity. Patients treated only with heparin (group 6) show an FPAvalue of 19 ng/ml, which is lower than in the other groups. All of this indicates that thrombin is activated after fibrinolytic treatment. FDP values show asignificant increase in the six groups (1, 53; 2, 430; 3, 128; 4, 270; 5, 139 and 6, 141 ug/ml), which indicates that during treatment with heparin the fibrinolytic activity persists. he formation of highly cross-linked fibrin is altered in groups 1,2,3 and 4,as a consequence of circulating FDP effect and fibrincgeno- lysis.The permeability of the fibrin clotdecreases in groups 1 (0.42), 2 (1.3), 4 (1.1) and 5(0.5 ml/s/ng) and increases in group 2 (23.2 ml/s/nig) with respect to the normal plasma value (3.2 ml/s/nrg). The decrease in permeability must be related to the existence of hypercoagulability resistant to heparinization. FPA values, tine Fgl/Fgll ratio, andfibrin permeability can be used to evaluate the degree of thrcmbin activity during thrombolytic treatmentinAMI.

Styles APA, Harvard, Vancouver, ISO, etc.

Nous offrons des réductions sur tous les plans premium pour les auteurs dont les œuvres sont incluses dans des sélections littéraires thématiques. Contactez-nous pour obtenir un code promo unique!