Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 4 of 4 for “"Gradient Inversion"”.
-
Images in motion?: a first look into video leakage in federated learning
… only exchange model updates, such as gradients, and never their raw, sensitive data. This approach is fundamental for applications in domains where privacy and confidentiality are important. However, the security of this very mechanism is threatened by gradient inversion attacks, which …
-
Understanding and measuring flow in aortic stenosis with MRI
… the duration of the velocity sensitivity gradients and shortening the overall echo time (TE). However, improvements in an optimised PC sequence (TE 1:5ms) were not satisfactory. Flow estimates remained variable and were underestimated beyond the aortic valve. To reduce the TE further, a …
-
Breaking Privacy in Model-Heterogeneous Federated Learning
… instead, clients only share locally computed gradients or model parameters with a central server. As individual gradients may leak information about a given client's dataset, secure aggregation was proposed. With secure aggregation, the server only receives the aggregate gradient update from …
-
Improving federated learning performance with similarity guided feature extraction and pruning.
… Perceptrons (MLP) trained using Stochastic Gradient Descent (SGD). First, we introduce FedSim, a similarity-guided model aggregation algorithm designed to enhance FL accuracy by leveraging inter-client relationships. Instead of relying on client data, we extract similarity knowledge by …