Scholarlink

In the last part of times, Federated learning (FL) has been presented as an approach for Artificial Intelligence (AI) in which models can be trained in a decentralized manner without violating users’ privacy. The data transmission and storage in the central server is one of the security risks of the traditional centralized learning methods. In this paper, this review paper look into the progress of FL in cloud AI with the potential to increase data privacy and efficiency as well as security. Additionally, we compare FL with the standard AI approach from the point of view of both accuracy and latency, as well as security resilience.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Federated Learning in Cloud AI: Enhancing Privacy and Security

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Raviteja Guntupalli

摘要

In the last part of times, Federated learning (FL) has been presented as an approach for Artificial Intelligence (AI) in which models can be trained in a decentralized manner without violating users’ privacy. The data transmission and storage in the central server is one of the security risks of the traditional centralized learning methods. In this paper, this review paper look into the progress of FL in cloud AI with the potential to increase data privacy and efficiency as well as security. Additionally, we compare FL with the standard AI approach from the point of view of both accuracy and latency, as well as security resilience.