About Federated Learning for Pervasive Systems
About this Collection
With the surge of data volumes generated by an increasing number of connected devices at the edge of the internet, coupled with the rise of privacy consciousness, the entire artificial intelligence process is taking a shift towards edge-centric computations instead of the traditional cloud approach. Thus, federated learning (FL) is now being actively explored. FL is a distributed learning framework that allows models to be trained and collaboratively learned across multiple computing devices and servers that store local datasets, without the need for data exchange. FL framework allows end-users to become co-creators of AI solutions while respecting their privacy. This approach has its own unique challenges, including dealing with data heterogeneity, integration, transparency, fairness, and network reliability. In addition, FL must also deal with the trustworthiness of edge devices with varying degrees of connectivity and processing power.
The aim of this collection is to promote the latest work regarding FL approaches, which are applied to novel applications of next-generation pervasive systems. We welcome submissions of research articles, case studies and systematic reviews which involve, but are not limited to, the development and/or application of distributed learning methods that address existing challenges. These challenges include privacy and security mechanisms, trustable model-centric sharing, and effective management techniques for real-world edge-centric applications.
Keywords: machine learning; distributed learning; fog networks; federated algorithms; privacy-aware algorithms; model aggregation; collaborative learning; deep learning; on-device learning; continual learning in pervasive systems; federated learning; personalised machine learning; optimisation; fairness
Any questions about this collection? Please get in contact directly with Kirsten Barr (Kirsten.Barr@F1000.com)
This collection is part of the Gateway on Artificial Intelligence & Machine Learning, which aims to provide stakeholders across academia, industry and policy with a space to disseminate work related to all areas of machine learning and AI research.
This collection is now closed
The aim of this collection is to promote the latest work regarding FL approaches, which are applied to novel applications of next-generation pervasive systems. We welcome submissions of research articles, case studies and systematic reviews which involve, but are not limited to, the development and/or application of distributed learning methods that address existing challenges. These challenges include privacy and security mechanisms, trustable model-centric sharing, and effective management techniques for real-world edge-centric applications.
Keywords: machine learning; distributed learning; fog networks; federated algorithms; privacy-aware algorithms; model aggregation; collaborative learning; deep learning; on-device learning; continual learning in pervasive systems; federated learning; personalised machine learning; optimisation; fairness
Any questions about this collection? Please get in contact directly with Kirsten Barr (Kirsten.Barr@F1000.com)
This collection is part of the Gateway on Artificial Intelligence & Machine Learning, which aims to provide stakeholders across academia, industry and policy with a space to disseminate work related to all areas of machine learning and AI research.
This collection is now closed