Sökning: "Dataintegritet"

Hittade 5 avhandlingar innehållade ordet Dataintegritet.

  1. 1. Pointwise Maximal Leakage : Robust, Flexible and Explainable Privacy

    Författare :Sara Saeidian; Tobias J. Oechtering; Mikael Skoglund; Giulia Cervia; Catuscia Palamidessi; KTH; []
    Nyckelord :TEKNIK OCH TEKNOLOGIER; ENGINEERING AND TECHNOLOGY; NATURVETENSKAP; NATURAL SCIENCES; Privacy; information leakage; pointwise maximal leakage; disclosure prevention; inferential privacy; mechanism design.; Dataintegritet; informationsläckage; punktvist maximalt läckage; avslöjningsprevention; inferentiell dataintegritet; mekanismdesign.; Electrical Engineering; Elektro- och systemteknik;

    Sammanfattning : For several decades now, safeguarding sensitive information from disclosure has been a key focus in computer science and information theory. Especially, in the past two decades, the subject of privacy has received significant attention due to the widespread collection and processing of data in various facets of society. LÄS MER

  3. 2. An Information-Theoretic Approach to Generalization Theory

    Författare :Borja Rodríguez Gálvez; Mikael Skoglund; Ragnar Thobaben; Benjamin Guedj; KTH; []
    Nyckelord :NATURVETENSKAP; NATURAL SCIENCES; Generalization; Information-Theoretic Bounds; Electrical Engineering; Elektro- och systemteknik;

    Sammanfattning : In this thesis, we investigate the in-distribution generalization of machine learning algorithms, focusing on establishing rigorous upper bounds on the generalization error. We depart from traditional complexity-based approaches by introducing and analyzing information-theoretic bounds that quantify the dependence between a learning algorithm and the training data. LÄS MER

  4. 3. Towards Realistic Smart Meter Privacy against Bayesian Inference

    Författare :Ramana Reddy Avula; Tobias J. Oechtering; Daniel Månsson; Mustafa A. Mustafa; KTH; []
    Nyckelord :TEKNIK OCH TEKNOLOGIER; ENGINEERING AND TECHNOLOGY; Smart meter privacy; Bayesian inference control; Energy storage model; Data-driven modeling; Privacy-enhancing mechanisms; Hidden Markov models; Co-living household energy dataset; Integritet för smarta elmätare; bayesiansk slutledningskontroll; energilagringsmodell; datadriven modellering; sekretessförbättrande mekanismer; dolda Markov-modeller; samlevande energidataset för hushåll; Electrical Engineering; Elektro- och systemteknik;

    Sammanfattning : Smart meters, now an essential component of modern power grids, allow energy providers to remotely monitor users' energy consumption in near real-time. While this technology offers numerous advantages for energy management and system efficiency, it also poses significant privacy concerns. LÄS MER

  5. 4. Application of Blockchain Technology for ISA95-Compliant Traditional and Smart Manufacturing Systems

    Författare :Erkan Yalcinkaya; Antonio Maffei; Mauro Onori; Primož Podržaj; KTH; []
    Nyckelord :TEKNIK OCH TEKNOLOGIER; ENGINEERING AND TECHNOLOGY; ISA95 standard; smart manufacturing; cybersecurity; interoperability; data quality; scalability; blockchain; blockchain system reference architecture; smart contracts; automation; machine to machine communication; industry 4.0; industrial internet of things; manufacturing industry;

    Sammanfattning : The ISA95 standard has been dominating the manufacturing industry for several years. A vast number of ISA95-compliant traditional and legacy manufacturing systems (ISA95-CTS) are scattered across many factories worldwide. The technological advancements imposed by Industry 4. LÄS MER

  7. 5. Distributed and federated learning of support vector machines and applications

    Författare :Shirin Tavara; Alexander Schliep; Alexander Karlsson; Lili Jiang; Högskolan i Skövde; []
    Nyckelord :NATURVETENSKAP; NATURAL SCIENCES; TEKNIK OCH TEKNOLOGIER; ENGINEERING AND TECHNOLOGY; NATURVETENSKAP; NATURAL SCIENCES; Skövde Artificial Intelligence Lab SAIL ; Skövde Artificial Intelligence Lab SAIL ;

    Sammanfattning : Machine Learning (ML) has achieved remarkable success in solving classification, regression, and related problems over the past decade. In particular the exponential growth of digital data, makes using ML inevitable and necessary to exploit the wealth of information hidden inside the data. LÄS MER