Federated Learning Framework

Summary

It is a Privacy Preserving Federated Learning Framework

• For: India’s Distributed Cloud Infrastructure.
• Challenges:
  • Digital Distributed Data Protection Bill 2023.
  • Networking Constraints:
    • High Latency.
    • Intermittent Connectivity.
• How?
  • Integrate advanced FL algorithms designed for hybrid cloud-edge environments, like:
    • Adaptive aggregation.
    • Data compression.
• Evaluation Metrics:
  • Accuracy.
  • Latency.
  • Cost.
• Datasets:
  • Healthcare.
  • Financial.
• Outcome:
  • Practical Open-Source FL Toolkit.
  • Actionable Policy Recommendations.
• Contribution:
  • Responsible and Efficient adoption of AI in data sensitive sectors across India.

Research Outline

• Introduction to Federated Learning and its Relevance for Privacy-Preserving AI
• India’s Digital Landscape: Cloud Infrastructure and Network Characteristics
• Navigating India’s Data Protection Landscape: The Digital Personal Data Protection Bill 2023
• Proposed FL Framework Design for Hybrid Cloud-Edge Environments
• Benchmarking Methodology and Performance Evaluation
• Open-Source FL Toolkit and Policy Recommendations
• Conclusions and Future Directions