This website is based on the paper “The Internet of Federated Things (IoFT)”. This paper provides a vision for the future IoT system along with an indepth overview of current efforts towards realizing this vision. IoFT is based on one key advancement in IoT: The computational power at the edge device has tremendously increased. AI chips are rapibly infiltrating the market, phone today has comparable compute power as everyday use laptops and small computers attached to the edge device (such as a Raspberry Pi in manufacturing) are becoming common place in many industries. Through exploiting the compute power at the edge, the “cloud” will be substituted by the “crowd” where model training is brought to the edge, allowing IoT devices to collaboratively extract knowledge and build smart analytics/models while keeping their personal data stored locally. The underlying data-driven approach for model learning in IoFT is coined federated learning/analytics denoting a decentralized model learning paradigm.
The purpose of this website
As IoFT is still in its infancy phase, real-life datasets (in engineering, health sciences, etc..) are pressingly needed to fully explore the disruptive potential of IoFT. While few already exist, they are based on artificial examples, and the few non-artificial datasets are mostly focused on mobile applications. However, for IoFT to become a norm in different industries, real-life datasets with defining features of the underlying system are needed to unveil the potential challenges and opportunities faced within different domains. Only with a deep understanding of the underlying system and domain, one formulates the right analytics.
As a result this website was created to serve as central directory for IoFT-based datasets. It features brief descriptions of each dataset categorized by its respective field with a link to the repository (research lab website, GitHub account, papers, etc..) where the data is contained. Our hope is to provide a means for model validation within different domains for IoFT, encourage researchers to develop real-life datasets for IoFT, and help with the outreach and visibility of their datasets and corresponding papers.
To add your dataset repository please reach out to us at email@example.com
- A Global model: One model to fit all. The global model aims at capturing the commonalities and intrinsic relatedness across data from all devices to improve prediction and learning accuracy.
- A Personalized model: that personalizes and adapt the global model to data and external conditions from each device. This embodies the principle of multi-task learning, where each device retains its own model while borrowing strength across all IoFT devices.
- A Meta-learning model: that learns a global model which can quickly adapt to a new task with only a small amount of training samples and learning steps. This embodies the principle of “learning to learn fast,” where the goal of the global model is not to perform well on all tasks in expectation, instead to find a good initialization that can directly adapt to a specific task.
The IoFT paper ends by describing the vision and challenges of IoFT in reshaping different industries through the lens of domain experts. Those industries include manufacturing, transportation, energy, healthcare, quality & reliability, business, and computing.