References of "Hofmann, Peter"
     in
Bookmark and Share    
Full Text
Peer Reviewed
See detailFairness, integrity, and privacy in a scalable blockchain-based federated learning system
Rückel, Timon; Sedlmeir, Johannes UL; Hofmann, Peter

in Computer Networks (2022), 202

Federated machine learning (FL) allows to collectively train models on sensitive data as only the clients’ models and not their training data need to be shared. However, despite the attention that ... [more ▼]

Federated machine learning (FL) allows to collectively train models on sensitive data as only the clients’ models and not their training data need to be shared. However, despite the attention that research on FL has drawn, the concept still lacks broad adoption in practice. One of the key reasons is the great challenge to implement FL systems that simultaneously achieve fairness, integrity, and privacy preservation for all participating clients. To contribute to solving this issue, our paper suggests a FL system that incorporates blockchain technology, local differential privacy, and zero-knowledge proofs. Our implementation of a proof-of-concept with multiple linear regressions illustrates that these state-of-the-art technologies can be combined to a FL system that aligns economic incentives, trust, and confidentiality requirements in a scalable and transparent system. [less ▲]

Detailed reference viewed: 133 (0 UL)
Full Text
Peer Reviewed
See detailHow ill is your IT Portfolio? : Measuring Criticality in IT Portfolios Using Epidemiology
Guggenmos, Florian; Hofmann, Peter; Fridgen, Gilbert UL

in 40th International Conference on Information Systems, Munich, Germany, 2019 (2019)

IT project portfolios, consisting of IT projects, also interact with the entire IT landscape. In case of a failure of only one element, existing dependencies can lead to a cascade failure, which can cause ... [more ▼]

IT project portfolios, consisting of IT projects, also interact with the entire IT landscape. In case of a failure of only one element, existing dependencies can lead to a cascade failure, which can cause high losses. Despite the present effects of systemic risk, research into IT portfolio management lacks suitable methods to quantitatively assess systemic risk. We follow the design science research paradigm to develop and evaluate our ‘on track’ or ‘in difficulty’ (TD) method by applying the SI model, representing a recognized network diffusion model in epidemiology, in an IT portfolio context. We evaluate our method using a real-world dataset. We introduce a criticality measure for diffusion models in IT portfolios and compare the TD method’s results and the alpha centrality to human judgment as a benchmark. From our evaluation, we conclude that the TD method outperforms alpha centrality and is a suitable risk measure in IT portfolio management. [less ▲]

Detailed reference viewed: 139 (1 UL)