Towards an Explainable, Efficient, and Reliable Federated Learning Framework: A Solution for Data Heterogeneity

• Project/Area Number: 24K20848
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 61030:Intelligent informatics-related
• Research Institution: Institute of Physical and Chemical Research
• Principal Investigator: HUANG Wei 国立研究開発法人理化学研究所, 革新知能統合研究センター, 研究員 (60976299)
• Project Period (FY): 2024-04-01 – 2027-03-31
• Project Status: Granted (Fiscal Year 2024)
• Budget Amount: ¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000); Fiscal Year 2026: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000); Fiscal Year 2025: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000); Fiscal Year 2024: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
• Keywords: Federated Learning / Data Heterogeneity / Deep Learning Theory

Outline of Research at the Start

Our goal is to develop an explainable, efficient, and reliable Federated Learning
Framework that will provide robust solutions to data heterogeneity. In effect, this enhances federated learning's utility and robustness in diverse real-world applications.