We are pleased to invite students, researchers, and industry professionals to a research seminar by Kuniko Paxton, Postdoctoral Edge AI Researcher at the University of Hull (UK) and member of the National Edge AI Hub.
Skewness-Guided Pruning of Multimodal Swin Transformers for Federated Skin Lesion Classification on Edge Device
📅 July 8, 2026
🕧 12:30 PM – 1:30 PM
📍 Bit Alliance Lab (3-46), 3rd Floor, Faculty of Electrical Engineering University of Sarajevo
Modern AI models are achieving remarkable success in medical image analysis, but their size and computational requirements often prevent deployment on resource-constrained devices. At the same time, privacy regulations make centralized collection of sensitive medical data increasingly difficult.
In this seminar, Kuniko Paxton will present a novel approach that combines model compression with privacy-preserving federated learning, making advanced AI more practical for deployment on edge devices.
Topics include:
🔹 Edge AI for medical imaging
🔹 Federated Learning for privacy-preserving AI
🔹 Vision Transformers and multimodal learning
🔹 Skewness-guided pruning of Swin Transformers
🔹 Efficient AI for skin lesion classification
Whether you are interested in Artificial Intelligence, Deep Learning, Computer Vision, Medical AI, Edge Computing, or Federated Learning, this seminar offers valuable insights into one of the most active research areas in trustworthy and efficient AI.
Participation is free. We look forward to welcoming you!
Kuniko_Paxton_seminar
