pfl research

We have hosted the application pfl research in order to run this application in our online workstations with Wine or directly.

Quick description about pfl research:

A fast, modular Python framework released by Apple for privacy-preserving federated learning (PFL) simulation. Integrates with TensorFlow, PyTorch, and classical ML, and offers high-speed distributed simulation (7–72× faster than alternatives).

Features:

• Simulates federated privacy-aware learning workflows
• Compatible with TensorFlow, PyTorch, scikit-learn
• Scales across processes, GPUs, multi-machine (via Horovod)
• Modular design for plugging privacy algorithms
• Benchmark suite for standardized comparisons
• Actively maintained by Apple researchers

Programming Language: Python.
Categories: