404
+Sorry, we couldn’t find the page you’re looking for. +
+Explore the NVIDIA FLARE Documentation. +
+Explore the NVIDIA FLARE GitHub. +
+NVIDIA FLARE is designed as a federated computing platform that is agnostic to frameworks, workloads, datasets, and domains. + View the Tutorial Catalog to see different examples in these categories. +
+NVIDIA FLARE is compatible with any machine learning or deep learning framework. + This versatility is demonstrated through various example repositories, including those for PyTorch, TensorFlow, PyTorch Lightning, XGBoost, Scikit-learn, GraphSAGE, Hugging Face, NeMo, Bio-Nemo, and MONAI. + Most machine learning problems can easily be converted from centralized algorithms to federated algorithms using NVIDIA FLARE. +
+NVIDIA FLARE is also domain agnostic, making it suitable for applications in medical imaging, drug discovery, self-driving cars, financial services, medical devices, energy, and more. + This broad applicability is reflected in the diverse industries of our customers. +
+NVIDIA FLARE supports a wide variety of models, including decision trees, classification, regression, various types of LLM fine-tuning, and XGBoost. + Regardless of the model type, NVIDIA FLARE can work with them all. +
+Thanks to its generic design, NVIDIA FLARE is task agnostic. + It can handle any type of task or payload, without mandating a specific type of computation. + NVIDIA FLARE provides coordination and communication, allowing users to leverage its capabilities for tasks such as Federated Statistics, Multi-Party Private Set Intersection (PSI), and Federated Retrieval-Augmented Generation (RAG) tasks. +
+Explore our extensive catalog of tutorials of NVIDIA FLARE features and examples for various applications. + Use the tags to filter by categories such as level, applications, frameworks, etc. + or search for keywords. + + To read more, visit our +documentation and +GitHub. +
+NVIDIA FLARE™ (NVIDIA Federated Learning Application Runtime + Environment) is a domain-agnostic, open-source, and extensible SDK + for Federated Learning. It allows researchers and data scientists + to adapt existing ML/DL workflow to a federated paradigm and + enables platform developers to build a secure, privacy-preserving + offering for a distributed multi-party collaboration. +
+Begin your NVIDIA FLARE journey with these guides + designed to help you quickly grasp essential concepts. + Follow along with the videos below, and try it out yourself. +
+Try out these example code blocks below, where we showcase how simple it is to adapt + a popular machine learning framework to a federated learning scenario with NVIDIA FLARE. + For more details, refer to the getting started walkthrough guide above. +
+These progressive tutorial series cover various aspects of NVIDIA FLARE from core concepts and basic tools to advanced algorithms and deployments. + For a comprehensive list of all tutorials, view the tutorial catalog below. +
Provides user-friendly APIs for client and server programming. Use the Simulator and POC modes to quickly simulate a federated learning application.
Designed as a federated computing platform agnostic to frameworks, workloads, datasets, and domains. Federated learning apps are built on this foundation.
Flexible open architecture allows for extensive customization, with a modular design that ensures each layer can be easily pluggable with custom components.
Prioritizes security with secure provisioning, event-based security plugins, authorization control, data filtering, audit logs, and advanced privacy-preserving algorithms.
Enables a seamless integration with external 3rd-party systems with the FlareAgent to receive tasks and submit results to the NVIDIA FLARE server.
Supports cross-cloud deployment with various management tools. Designed for robust, production-scale deployment in real-world federated learning scenarios.
Federated Learning is a distributed learning paradigm where training occurs across multiple clients, each with their own local datasets. This enables the creation of common robust models without sharing sensitive local data, helping solve issues of data privacy and security.
NVIDIA Federated Learning Technical Blogs covering various applications and topics.
List of NVIDIA FLARE related papers and publications along with videos from blogs and talks.
+Learn how NVIDIA FLARE faciliates research from simulation tools to real-world case studies. +
+NVIDIA FLARE is an excellent research tool, offering robust simulation capabilities and extensive support for experimentation in federated learning. + Many researchers need to simulate federated learning scenarios without setting up an actual federated learning system. + NVIDIA FLARE allows for repeated experimentation with different parameters, facilitating quick evaluations and monitoring of results. +
Simulator: + The Simulator is a multi-threaded/process simulation tool that offers both a Command Line Interface (CLI) and a Python API. + It enables the simulation of different numbers of clients and the execution of various federated learning jobs. + Once a simulation is complete, users can deploy the same code in production without any changes. + Additionally, users can utilize an Integrated Development Environment (IDE) debugger to step through the code for easier debugging. +
Proof of Concept (POC) Mode: + POC mode simulates real-world deployment on a local host. + Clients and servers can be deployed in different directories and launched using separate terminals, each representing a different client or server startup. + This mode allows for job submissions to the server as would occur in a real production environment. +
Admin Console: + Issue interactive commands such as submitting jobs, listing jobs, and aborting jobs. +
Job CLI: + Command-line interface is used for job submission. +
FLARE API: + Allows submission and listing of jobs through Python code. +
Experiment tracking: + FLARE supports logging metrics using FLARE's metrics tracking log writers. Users can choose from TensorBoard, MLflow, or W&B syntax. + Metrics can be streamed to either the FL server or FL client, and changing the metric system does not require any code changes. +
+NVIDIA FLARE offers a lot of state of art research work, here is a quick view of the recent work. + Learn more in the research directory in our GitHub and view our list of Publications. +
+NVIDIA has worked with several institutions to test and validate the utility of federated learning. We hope this will help to inspire or relate to your research cases.
+
+Here are five real life implementations in healthcare, pushing the envelope for training robust, generalizable AI models:
+
AI model to predict oxygen requirements
NVIDIA researchers, Massachusetts General Brigham Hospital
Breast Mammography AI Model. Early detection: breast density classification Improvement
The American College of Radiology (ACR), Diagnosticos da America (DASA), Ohio State University (OSU), Partners HealthCare (PHS), and Stanford University
Improve AI models for COVID-19 diagnosis based on chest X-rays
University of Minnesota and Fairview Mhealth, Indiana University (Indiana, USA), and Emory University (Georgia, USA)
Federated segmentation model
SUNY, UCLA, NIH
Automated segmentation model of the pancreas and pancreatic tumors in abdominal CT
National Taiwan University, Taiwan, and Nagoya University, Japan
+NVIDIA FLARE implements comprehensive security measures across various areas. +
Communication Security:
+ Ensures the confidentiality of data communication messages.
+
+To learn more, visit the Security section in our documention.
+