Call for Contributions: NVIDIA FLARE SUMMER OF CODE 2025 #3238
holgerroth
announced in
Announcements
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Join the Effort!
Are you passionate about advancing federated learning? Contribute to NVIDIA FLARE, collaborate with the open-source community, and help shape the future of privacy-preserving and scalable AI!
How to Get Involved
Why Participate?
Key Dates
Focus Areas
We welcome contributions in the following areas:
🔹 Energy-Efficient Communication & Learning
(e.g., compression, pruning, parameter-efficient finetuning, asynchronous FL, carbon footprint tracking)
🔹 Privacy-Preserving Techniques in the Age of LLMs
(e.g., differential privacy, memorization prevention, privacy attacks & defenses)
🔹 Training on the Edge (Real-Time & Multi-Sensor Data)
(e.g., mobile, automotive, robotics, surgery, dynamic sensor networks)
🔹 Collaborative Inference & Information Aggregation
(e.g., federated RAG, multi-agent learning, decentralized inference)
What Contributions Are We Looking For?
✅ Algorithms – Novel methods to improve federated learning efficiency, scalability, and security
✅ Core Framework & Utilities – Enhancements to NVFlare’s core functionalities
✅ Examples & Use Cases – Demonstrations of FL applications in real-world scenarios
Ready to contribute? Join us and make an impact!
Submission Guidelines
To participate, follow these steps:
1️⃣ Register a Proposal
2️⃣ Develop Your Contribution
3️⃣ Submit a Pull Request (PR)
4️⃣ Review & Iterate
5️⃣ Final Submission & Selection
By following this structured approach, your solution will have a greater impact and visibility in the federated learning community!
Feel free to ask any questions you might have here.
Beta Was this translation helpful? Give feedback.
All reactions