Federated Machine Learning for Intelligent IoT via Reconfigurable Intelligent Surface

Blockchain for federated learning toward secure distributed machine learning systems: a systemic survey

“…Currently, FL has been integrated with other emerging technologies by many scholars to enable industrial applications, such as the efficiency improvement of mobile and wireless communication (Konecný et al 2016;Sattler et al 2020;Reisizadeh et al 2020;Niknam et al 2020), edge computing Doku et al 2021;Fantacci and Picano 2020;, health care (Rieke et al 2020;Bogdanova et al 2020;Zerka et al 2020), Internet of Things (Savazzi et al 2020;Yang et al 2020;Yuan et al 2020;Qolomany et al 2020;Briggs et al 2020;Gao et al 2020;Kamel and Mougy 2020;Imteaj and Amini 2019), Internet of Vehicles (Samarakoon et al, 2020;Hsu et al, 2020;, anomaly detection (Nguyen et al 2019;Weinger et al 2020), smart city (Jiang et al 2020), financial fraud identification (Fan et al 2020), visual object detection ) and fog computing ). It can be seen that FL is prominent in industrial applications for privacy-sensitive data and the processing of non-IID data.…”

Artificial intelligence techniques in finance and financial markets: A survey of the literature

“…Within these technologies, AI offers continuous innovation for the improvement and potential replacement of human tasks and activities with a wide range of applications in finance, healthcare, manufacturing, retail, supply chain, logistics, and utilities (Ben‐Israel et al, 2020; Bredt, 2019; Dwivedi et al, 2019; Safdar, Banja, & Meltzer, 2020). The internet of things (IOT) provides vast amounts of data that would lead to connected intelligence, enabling applications such as self‐driving cars, uncrewed aerial vehicles, healthcare, robotics, and supply chain finance (Yang et al, 2020).…”

“…Similarly, the IoT, ML, and AI are helping supply chain management by regrouping data at diverse places and increasing the efficiency of supply chain finance (McCrea, 2019). The use of massive data afforded by IoT needs to be weighed with each user's privacy and security needs, indicating that intelligence needs to be separated from the data, and this is being done by federated ML (Yang et al, 2020). Federated learning or collaborative learning allows learning from several sources without exchanging information.…”

AI-Assisted MAC for Reconfigurable Intelligent-Surface-Aided Wireless Networks: Challenges and Opportunities

“…The centralized learning employed for the MAC design of RIS-aided communications may rely on privacy-sensitive datasets. To address this potential vulnerability, distributed learning, e.g., federated learning [14], [15], requires uploading of only local model parameters, thereby promoting data privacy, although steps must be taken to preserve privacy even with distributed learning. Moreover, privacy can be enhanced in both training and inference through the use of federated learning and neural network segmentation, respectively.…”