432 citations
- Centre National de la Recherche ScientifiqueFR5 papers
- Istanbul Technical UniversityTR5 papers
- University of WolverhamptonGB4 papers
- Helmholtz-Zentrum Berlin für Materialien und EnergieDE3 papers
- Observatoire de ParisFR3 papers
- Queen's University BelfastGB3 papers
- Shanghai Ocean UniversityCN3 papers
- University of GlasgowGB3 papers
- Chinese Academy of SciencesCN2 papers
- Fudan UniversityCN2 papers
- Janssen (France)FR2 papers
- Laboratoire Univers et ThéoriesFR2 papers
6 papers · 2 filters
Launching Adversarial Attacks against Network Intrusion Detection Systems for IoT
Pavlos Papadopoulos, Oliver Thornewill von Essen, Nikolaos Pitropakis +3
As the internet continues to be populated with new devices and emerging technologies, the attack surface grows exponentially. Technology is shifting towards a profit-driven Interne…
Practical Defences Against Model Inversion Attacks for Split Neural Networks
Tom Titcombe, Adam J. Hall, Pavlos Papadopoulos +1
We describe a threat model under which a split network-based federated learning system is susceptible to a model inversion attack by a malicious computational server. We demonstrat…
PyVertical: A Vertical Federated Learning Framework for Multi-headed SplitNN
Daniele Romanini, Adam James Hall, Pavlos Papadopoulos +6
We introduce PyVertical, a framework supporting vertical federated learning using split neural networks. The proposed framework allows a data scientist to train neural networks on…
FairDrop: Biased Edge Dropout for Enhancing Fairness in Graph Representation Learning
Indro Spinelli, Simone Scardapane, Amir Hussain +1
Graph representation learning has become a ubiquitous component in many scenarios, ranging from social network analysis to energy forecasting in smart grids. In several application…
FastAdaBelief: Improving Convergence Rate for Belief-based Adaptive Optimizers by Exploiting Strong Convexity
Yangfan Zhou, Kaizhu Huang, Cheng Cheng +3
AdaBelief, one of the current best optimizers, demonstrates superior generalization ability compared to the popular Adam algorithm by viewing the exponential moving average of obse…
A New Class of Efficient Adaptive Filters for Online Nonlinear Modeling
Danilo Comminiello, Alireza Nezamdoust, Simone Scardapane +3
Nonlinear models are known to provide excellent performance in real-world applications that often operate in non-ideal conditions. However, such applications often require online p…