A Night by Immeln Lake: Rediscovering Wilderness in Southern Sweden
Explore the quiet wilderness of Immeln Lake and the Brotorpet nature reserve, where the old woods and vibrant wildlife come alive.
Artificial Mind
Explore this AI enthusiast's blog and resume, documenting a dedicated journey in AI and machine learning, enriched with personal experiences, accomplishments, and captivating photographies.
Pont Royal
Møns Klint: Denmark's Majestic White Cliffs in Spring
Experience the natural splendor of Møns Klint as spring breathes new life into Denmark's iconic white cliffs.
FLEX: Advancing Federated Learning Research with a Versatile Framework
Introducing FLEX, a comprehensive framework that addresses key challenges in federated learning simulation and enables innovative research across various machine learning domains.
PlateSegFL: Enhancing License Plate Detection with Privacy-Preserving Federated Learning
A new approach combining U-Net segmentation and Federated Learning achieves accurate license plate detection while preserving user privacy, addressing challenges in irregular masking and data security.
Improving SecureBoost: Balancing Performance, Efficiency, and Privacy in Federated Learning
A new approach called Constrained Multi-Objective SecureBoost (CMOSB) optimizes hyperparameters to enhance model performance, reduce training costs, and strengthen privacy protection in federated learning.
Enhancing Federated Learning Security: A Consensus-Based Approach to Model Update Validation
Explore an innovative method for protecting Federated Learning systems against label-flipping attacks using consensus-based verification and adaptive thresholding techniques.
Federated Learning and the EU AI Act: A Balancing Act
A new paper explores how federated learning can adapt to align with the EU AI Act's principles on privacy, robustness, and energy efficiency.
FRECA: A Fair, Robust, and Efficient Method for Evaluating Client Contributions in Federated Learning
Learn about FRECA, a new approach for assessing client contributions in federated learning that ensures fairness, defends against attacks, and minimizes computational costs.
Learn2pFed: Adaptive Partial Parameter Collaboration in Personalized Federated Learning
Learn2pFed enables clients to adaptively select local model parameters for federated collaboration, improving performance in personalized federated learning by addressing data heterogeneity.