68 Hits in 12.2 sec

Quality Inference in Federated Learning with Secure Aggregation [article]

Balázs Pejó and András Tótth and Gergely Biczók
2022 arXiv   pre-print
Federated learning algorithms are developed both for efficiency reasons and to ensure the privacy and confidentiality of personal and business data, respectively.  ...  Hence, secure aggregation is utilized in many real-world scenarios to prevent attribution to specific participants.  ...  A disaggregation technique is presented in [44] which reconstructs the participation matrix by simulating the same round several times with different participants.  ... 
arXiv:2007.06236v3 fatcat:yemokykjtza3hf5ovmnzzg3mhu

Eluding Secure Aggregation in Federated Learning via Model Inconsistency [article]

Dario Pasquini, Danilo Francati, Giuseppe Ateniese
2022 arXiv   pre-print
We devise two different attacks capable of inferring information on individual private training datasets, independently of the number of users participating in the secure aggregation.  ...  Indeed, the use of secure aggregation prevents the server from learning the value and the source of the individual model updates provided by the users, hampering inference and data attribution attacks.  ...  This disaggregation process can be reduced to a matrix factorization problem.  ... 
arXiv:2111.07380v4 fatcat:l4ehhuu5xzfbvkfgv7mzuaxlra

D1.1 - State of the Art Analysis

Danilo Ardagna
2021 Zenodo  
Then, the deliverable provides a background on AI applications design, also considering some advanced design trends (e.g., Network Architecture Search, Federated Learning, Deep Neural Networks partitioning  ...  In the last part of the deliverable, we report an overview of the performance modelling solutions, security, and privacy problems for AI applications in edge environments.  ...  : an attacker, as shown in Figure 6 .2, with access to the sharing gradients party, could exploit the information contained in the models updates to reconstruct the training data by using a Generative  ... 
doi:10.5281/zenodo.6372377 fatcat:f6ldfuwivbcltew4smiiwphfty

Deep Neural Mobile Networking [article]

Chaoyun Zhang
2020 arXiv   pre-print
This thesis attacks important problems in the mobile networking area from various perspectives by harnessing recent advances in deep neural networks.  ...  The performance of artificial intelligence (AI) has achieved in other domains draws unprecedented interest from both academia and industry in employing deep learning approaches to address technical challenges  ...  Initial research is conducted in [126] , where the authors enable user participation in the training and evaluation of a neural network, without sharing their input data.  ... 
arXiv:2011.05267v1 fatcat:yz2zp5hplzfy7h5kptmho7mbhe

Technical Program

2022 2022 IEEE International Conference on Consumer Electronics (ICCE)  
Simulation results show that the proposed method can adapt to both ambient light levels and the scene content to keep the creative intent in different ambient conditions.  ...  The paper proposes a high dynamic range (HDR) tone mastering system which dynamically corrects the picture quality based on creative intent metadata to preserve content providers' creative intent in different  ...  We use the camera to capture the image data when the user is signing in the air. In addition, the trajectory of the finger is captured by using deep learning model on the embedded system.  ... 
doi:10.1109/icce53296.2022.9730380 fatcat:csqu3xqbczgdhpp3hbmvjpt26a

Introduction – The Contributions of Big Data
Introduction – Les apports des Big Data

Philippe Tassi
2019 Economics and Statistics  
They are also a driver for development in computer science in the broadest sense, and in applied mathematics.  ...  This Special Issue features a series of articles that examine its uses and implications, as well as the challenges faced by statistical production in general, and especially that of official statistics  ...  In high dimension, the numerical descent and gradient methods may be used, where the gradient is approximated by subsampling (Zinkevich et al., 2010) .  ... 
doi:10.24187/ecostat.2018.505d.1963 fatcat:42rehacy6ffb7hc6k2l2q4u4yy

WHISPER: Wireless Home Identification and Sensing Platform for Energy Reduction

Margarite Jacoby, Sin Yong Tan, Mohamad Katanbaf, Ali Saffari, Homagni Saha, Zerina Kapetanovic, Jasmine Garland, Anthony Florita, Gregor Henze, Soumik Sarkar, Joshua Smith
2021 Journal of Sensor and Actuator Networks  
We present the WHISPER (Wireless Home Identification and Sensing Platform for Energy Reduction) system to address the energy and power demand triggered by human presence in homes.  ...  Several machine learning algorithms are implemented at the base station to infer human presence based on the received data, harnessing a hierarchical sensor fusion algorithm.  ...  can lead to a multitude of privacy concerns for the users [35] .  ... 
doi:10.3390/jsan10040071 fatcat:ldapbfvlxzb23jo62nadduwypy

Transport-domain applications of widely used data sources in the smart transportation: A survey [article]

Sina Dabiri, Kevin Heaslip
2018 arXiv   pre-print
introduced to fuse the knowledge learned from a set of heterogeneous but complementary data sources.  ...  Secondly, as the most salient feature of this study, the transport-domain applications of each data source that have been conducted by the previous studies are reviewed and classified into the main groups  ...  The promising goal of deep learning algorithms is to learn and extract image representation features automatically rather than using the hand-crafted features designed by a user.  ... 
arXiv:1803.10902v3 fatcat:tc67qy4x4vbtjb76qi6mbwrqy4

Big data and extreme-scale computing

M Asch, T Moore, R Badia, M Beck, P Beckman, T Bidot, F Bodin, F Cappello, A Choudhary, B de Supinski, E Deelman, J Dongarra (+27 others)
2018 The international journal of high performance computing applications  
introduced by the ongoing revolution in high-end data analysis (HDA) might be integrated with the established, simulation-centric paradigm of the high-performance computing (HPC) community.  ...  We close by offering some conclusions and recommendations for future investment and policy review.  ...  Acknowledgments The authors would like to acknowledge David Rogers for his work on the illustrations, Sam Crawford for editing support and the creation of the Appendix, and Piotr Luszczek for technical  ... 
doi:10.1177/1094342018778123 fatcat:vwrrxmad4rhtppq6ioaz4h5q7a

Spatiotemporal enabled Content-based Image Retrieval

Mariana Belgiu, Martin Sudmanns, Tiede Dirk, Andrea Baraldi, Stefan Lang
2016 International Conference on GIScience Short Paper Proceedings  
The total area covered by a WSN is obtained from the union of the regions covered by individual sensors.  ...  This is because raster representations are constrained by their spatial resolution, and their regular shapes result in redundant data for unoccupied areas.  ...  Future Work Future research will extend our machine learning approach on Spark in the following four directions: (1) scanning a wider parameterization space and further optimizing search methods for the  ... 
doi:10.21433/b311729295dw fatcat:fulw4pw3kfh5nmfzcsy3pkisvm

Multi-period portfolio selection with drawdown control

Peter Nystrup, Stephen Boyd, Erik Lindström, Henrik Madsen
2018 Annals of Operations Research  
The participants were negotiators, who took part in real life labour bargaining processes, as found by the list of the Organisation for Mediation and Arbitration of a European country.  ...  In order to obtain the desired relative accuracy prescribed by the user, we use the past accuracy of the forecasts at all levels to determine which ones to trust depending on their historical ability to  ...  We discuss the obtained results from the business perspective covering forecasts applications in the inventory optimization and demand planning areas.  ... 
doi:10.1007/s10479-018-2947-3 fatcat:haworusnrfcqvlmkmbidfwb5wu

High-quality health systems in the Sustainable Development Goals era: time for a revolution

Margaret E Kruk, Anna D Gage, Catherine Arsenault, Keely Jordan, Hannah H Leslie, Sanam Roder-DeWan, Olusoji Adeyi, Pierre Barker, Bernadette Daelmans, Svetlana V Doubova, Mike English, Ezequiel García Elorrio (+20 others)
2018 Lancet Global Health  
, and greater public participation.  ...  Additional targeted actions in areas such as health financing, management, district-level learning, and others can complement these efforts.  ...  Acknowledgments The work of this Commission was supported by a grant to MEK from the Bill & Melinda Gates Foundation.  ... 
doi:10.1016/s2214-109x(18)30386-3 pmid:30196093 pmcid:PMC7734391 fatcat:ejlbyetgx5alxbweqqjpomuqky

Comparative Analysis of Different Machine Learning Classifiers for the Prediction of Chronic Diseases [chapter]

Rajesh Singh, Anita Gehlot, Dharam Buddhi
2022 Comparative Analysis of Different Machine Learning Classifiers for the Prediction of Chronic Diseases  
A comparative study of different machine learning classifiers for chronic disease prediction viz Heart Disease & Diabetes Disease is done in this paper.  ...  This paper forms the basis of understanding the difficulty of the domain and the amount of efficiency achieved by the various methods recently.  ...  option for power generation in the commercial space such that the penetration of this technology into market would improve the energy efficiency and also quality of the environment by decarbonization.  ... 
doi:10.13052/rp-9788770227667 fatcat:da47mjbbyzfwnbpde7rgbrlppe

Real-Time Ridesharing

Andrew Amey, John Attanucci, Rabi Mishalani
2011 Transportation Research Record  
The trial is designed to be expanded to other institutions in the MIT/Kendall Square area of Cambridge, MA in the future.  ...  Unfortunately, ridesharing's historical success has been rather modest, with a substantial decrease in popularity since the 1970's and participation that remains near an all-time low.  ...  of incentives for rideshare participation must continue to be considered an important second-best strategy.  ... 
doi:10.3141/2217-13 fatcat:7zfsm2n2gveohoephlqyzdersm

Abstracts, Oral for Qualitative Methods Conference

2016 International Journal of Qualitative Methods  
In New Zealand, nearly one quarter of the gay and bisexual men newly diagnosed with HIV are of Asian ethnicity.  ...  An Exercise in Calibration Jaclyn Tejwani, WestEd Stephen Ruffini, WestEd Marycruz Diaz, WestEd Reino Makkonen, WestEd  ...  The study focuses on why housing competence is central to people's lives and how users coping with household challenges may be strengthened by user participation and task-centered coping approaches (TCCAs  ... 
doi:10.1177/1609406916628953 fatcat:zjby346txfgfdcc552suhppzla
« Previous Showing results 1 — 15 out of 68 results